ED332398.Pdf

DOCUMENT RESUME

ED 332 398 EC 232 547

AUTHOR Pendarvis, Edwina D.; And Others TITLE The Abilities of Gifted Children. REPORT NO ISBN-0-13-002072-9 PUB DATE 90 NOTE 400p. AVAILABLE FROMPrentice-Hall, Prentice Hall Bldg., Englewcod Cliffs, NJ 07632 ($36.95). PUB TYPE Books (010) -- Guides - Classroom Use - Teaching Guides (For Teacher) (052)

EDRS PRICE 14F01 Plus Postage. PC Not Available from EDRS. DESCRIPTORS *Academically Gifted; Adults; *Child Development; Creativity; Delivery Systems; *Educational Needs; Elementary Secondary Education; *Gifted; Gifted Disadvantaged; Intelligence; Leadership? Low Income Groups; Minority Groups; Nature Nurture Controversy; Student Educational Objectives; *Talent; Underachievement

ABSTRACT The book .!.,1 designed as the primary text for courses that consider the nature of giftedness. It offeh's a strong introduction to the study of gifted education, and is accessible to both undergraduate and graduate students. Chapters address the following topics: the study of gifted students; heredity ani environment; intelligence constructs; creativity constructs; adulthnodeminence, leadership, and careers; academically talented childrun; children with talent in the visual and performing arts; typical gifted development; precocious development; inhibited developmentunderachievement; unfair discrimination, poverty, and minority gifted children; and methods of talent development. Extensive references are included. (DB)

**********************************************************************

..t * Reproductions supplied by EDRS are the best that can be made * * from the original document. *********************************************************************** THE ABILITIES OF GIFTED CHILDREN

Oirsorrourr or EDUCATION Mice ot Eductomnai R011eirCh end Imenmemont EDUCATIONAL RESOURCES INFORMATION CENTER (EMC) /Inddocument has Men reproduced sa Edwina D. Pendarvis reClorCI tforn me penal or otcsmumon coveting d Marshall University r Wo' Chopre Mr* Orin mad* to unotove rOtOCIOChon qualify Aimee A. Howley Porn OS reve Or OO,fIiOI Meted in l hal (Wu- moot do not 'Womanly reOnment otticial University of Charleston MR; cmerton cootCy

Craig B. Howley "PERMISSION TO REPRODUCE THIS Appalac-hia Educational Laboratory MATERIAL IN MICROFICHE ONLY HAS ESN GRANTED

TO THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC)." BEST COPY AVAILABLE

AT% PRENTICE HALL, Englewood Cliffs, New Jersey07632 ft Library of Congress Cataloging-in-Publication Data 1tndarvis, F.divina 1). The abilities of gifted children /Edwina D. Pendarvis, Aimee Hawley, Craig Howley. cm. ipjfilLtp, 357 Ind ISBN 043-002072-9 1. Gifted children.I. Hawley, Aimee.11. Hawley, Craig B. 1C3993.P3111990 371.95dc20 89-$789 C1P

Editorial/production supervision and interior design: Shelly Kupperman Cover design: 20/20 Services Inc. Manufacturing buyer: Pete Havens

For our parents: Marionette and Edward Roberta and Aaron Lorna and Jim

0 1990 by Prentice-Haft, loc. A Division of Simon & Schuster Englewood Cliffs, New Jersey 07632

AU rights reserved. No part of this book nay be reproduced, in any form ar by any MIMS, without permission in writing front the publisher.

Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

ISBN 0-13-002072-9

Prentice-Hall International (UK) Limited, London Prentice-Hall of Australia Pty. Limited. Sydney Prentice-Hall Canada Inc., Toronto Prentice-Hall Hispanoamericana, S.A., Mexico Prentice-Hall of India Private Limited, New Delhi Prentice-Hall of Japan, Inc., Tokyo Simon & Schuster Asia Pte. Ltd., Singapore Editora Prentice-Hall do Brasil, Ltda., Rio de Janeiro

3 Contents _

PREFACEIX

Part I:Context

THE STUDY OF GIFTED STUDENTS 1 1 Focusing Questions2 What Is the Study of Gifted Students? 2 Definitions3 Functional Views of Giftedness5 Characteristics: Empirical Views of Giftedness 12 Interpreting Empirical Results14 Talent in the School Setting17 Summary 25 HEREDITY AND ENVIRONMENT 27 Focusing Questions 28 The Nature and Nurture of the Debate about Nature and Nurture28 Research on the Nature versus Nurture Question30 Historical Identification Trends 37 Misconceptions in Hereditarian Interpretations of Research41 Alternate Conclusions 42 Summary 43

lit iv Table of Coded' Part Kinds of Talent INTELLIGENCE CONSTRUCTS 44 3 Focusing Questions 45 Normal and Abnormal intelligence46 Adaptive Behavicr54 School Performance58 Intellect68 School Responses to High-IQ Children72 Summary 78 CREATIVITY CONSTRUCTS 79 Focusing Questions 80 Intelligence and Adult Productivity80 Issues of Convergence and Divergence 81 Personality Traits Associated with Creativity87 Creativity and Intelligence95 School Responses to Creativity102 Summary 105 ADULTHOOD: EMINENCE, LEADERSHIP, AND CAREERS 106 Focusing Questions107 Concern for the Adulthood of Gifted Students108 Definitions109 Eminence and Giftedness113 Leadership and Giftedness120 Occupations and Giftedness133 Summary 145 ACADEMICALLY TALENTED CHILDREN 146 Focusing Questions147 Relationship Between IQ and Academic Talent 148 Verbal Precocity156 Mathematical Ability167 Summary 178 CHILDREN WITH TALENT IN THE VISUAL AND PERFORMING ARTS 180 Focusing Questions181 Arts Talent, Creativity, and Academics182 Music as a Concern of Schooling 186 Talent in the Visual Arts: The Arts Context194 Table of Contexts

Dancing and Acting as Performing Arts201 Summary 211

Part III:Talent Development TYPICAL GIFTED DEVELOPMENT 212 Focusing Questions 213 Types and Stereotypes214 Developmental Domains: Types and Anomalies 219 Language Acquisition and Verbal PrecociV 219 Concept Development 227 Social and Emotional Development of Gifted Children233 Summary 240

PRECOCIOUS DEVELOPMENT 242 Focusing Questions243 Extreme Precocity 243 Children With Very High lQs 245 Extreme Mathematical Precocity247 Musical Precocity253 Literary Precocity258 Athletic Precocity260 Other Precocity261 Summary 264

INHIBITED DEVELOPMENT: UNDERACHIEVEMENT 265 Focusing Questions 266 Etiology of Underachievement 266 Characteristics of Gifted Underachievers274 Handicapped Gifted Children280 Summary 293

UNFAIR DISCRIMINATION, POVERTY, AND MINORITY GIFTED CHILDREN 294 Focusing Questions 295 Fact: Discrimination Exists295 Effects of Discrimination297 Minority Gifted Children301 The Relationship of Gender to Patterns of Underachievement 307 Gifted Students in Rural Schools313 Summary 319

t; vi Tabk of Cambrak Part IV:Implications METHODS OF TALENT 12 DEVELOPMENT 321 Focusing Questions 322 Identification Procedures 323 Placement Options and Procedures 327 Acceleration332 Enrichment 337 Curriculum: Academics and the Arts340 Characteristics of Instruction for Gifted Students 347 Summary 354

REFERENCES 357 INDEX 384 Preface

One of us was recently asked by a state agency to speak to a groupof teachers and administrators about the condusions of research onaccelera- tion. The talk took place at an annual statewide meeting, and the new state regulations, which for the first time specified this strategy by name, were on the agenda. At the end of the talk, however, an official of the agency rose tomake this statement: "Acceleration should be used with only a smallfraction of gifted students," This remark would have had the intended effect, had not an outspoken educator from oneof the universities conducting research on acceleration challenged the remark. According to this educator,the strategy was appropriate for85.percent of gifted students. This anecdote illustrates that supposition and sentiment, ratherthan informed judgment, too frequently guide the evolution ofschool programs. In the hectic climate of thedassroom, it is, in fact, not easy to consult research. Intuition necessarily plays a valuable role inhelping feathers make the countless decisions that each day demandsof them. Whereas this pattern of intuitive response makes in the press of classroom events, it does not make sense for plannirff titherlessons or programs. Planning for lessons and programs reqesknowledge more than intuition. Three types of knowledge apply: knriwledge ofsubject matter, knowledge of students, and knowledgeof teachng methods. There are no substitutes; neithersentiment nor supposition will do. We wrote this book to give teachers the second kindof knowledge, knowledge about what gifted children a-e like. In the lastdecade a great deal of research has examined this topic. Our goal was to make senseof it. First, we wanted to make sense of it for ourselves. Without aclear personal understanding, we could not make sense of it for others. Second, onthe basis of that understanding, we wanted to presentthe research in a way that would help teachers and administrators make rationaldecisions about bright str,dents. Not only did we think it important to bring togetherthe conclusions of the best research, we also saw the need to provide a contextfor inter- via Profact preting the findings. Such a context acknowledges both the value and the limitations of empi, ical research, for empirical research makes sense only in light of well-reasoned theory. The Abilities of Gyied Children is designed za the primary text for courses that consider the nature ofgiftedness. Some such courses emphasize the characteristics of gifted children, and others focus onthe psychological constructs that determine talent. This book is relevant toeither conception. In addition, it offers a strong introduction to thestudy of gifted education. Because the text is highly readable, it is accessible to both undergraduate and graduate students.

ACKNOWLEDGMENTS Writing a book is a selfish act. It deprives your children ofattention; it bores your friends; it puzzles your colleagues; andas your grass goes unmowedit must secretly irk your neighbors. Therefore, wethink it is fitting to express thanks to all of these people for theirforbearance. We also want to thank others for their help with themanuscript: Susan Wil lig and Carol Wada, our editors at PrenticeHall, for their sense of perspective and their optimism; ShirleyFerguson, for reading both of our textbooks back toback; Pat Cahape, the mother of two mathematically talented girls, for her review of Chapter 7; KathyShowen and the teachers of the gifted in Jackson County, West Virginia,for their enthusiasm and support; and the staff of theAppalachia Educational Laboratory, for their persistence in bringing research and practice together. Wegreatly ap,,reci- ate the helpful reviewer commentsand suggestions: Steven Christ . ?her- son, United States Air ForceAcademy; Dean K. McIntosh, University of South CarolinaColumbia; Madt Pius, SeattlePacific University; Avis J. Ruthven, Mississippi State University; Beverly D.Shaklee, Kent State Uni- versity; and Terry A. Thomas, California StateUniversitySacramento. Earlier versions of material in Chapter 5 and Chapter 1 Ihave appeared in the Journal for the Education of the Gifted andthe Rural Special Education Quarterly, and we wish to thank thesepublications for their permission to use this material here. This book has claimed three years of our lives. Its progresshas paral- leled our children's travels through variousschool experiences and our own progress throughvarious professional experiences. This background is a constant accompaniment to our writing;the experiences and the writing have taken meaning from each other. We hope youwill find this book's connections to your lifeas a student, teacher, parent,researcher, and writer.

f4 1 Tne Study of Gifted Students

1. Focusing Questions II. What Is the Study of Gifted Students? III. Definitions A.Theoretical Views of Giftedness 1. Social attributes 2.Psychological constructs IV. Functional Views of Giftedness A.Pedagogical Definitions of Giftedness 1, Federal definition 2.State definitions B. Instrumental Definitions 1.National security 2. Business definitions V. Characteristics: Empirical Views of Giftedness A.Research About Students' Characteristics 1. Case studies 2. Quantitative cross-sectional studies 3. Longitudinal studies VI. Interpreting Empirical Results A.Recognizing Assumptions 1. Focus on the individual 2. Faith in measurability 3.Faith in research as a guide to practice B.Applying Research About Characteristics VII. Talent in the School Setting A.Finding Talent 1. Rarity and universality

`. 1 1 ) The Study of Gifted &admix

2. Practical methods of identification 3.Error and equity B. Rewarding and Punishing Talent 1.Elitism 2.Anti-intellectualism C.Developing Talent 1. Gifted programs 2.School improvement VIII.Summary

Focusim Questions 'I.How do theoretical and Linctional views of giftedness differ? Which is more likely to result in a coherent definition? Why? 2.What role has business played to support the development of programs for gifted students? Which kind of ability receives this support, general intellectual ability or creative ability? 3.What are the differences among the various types of empirical studies of gifted students? 4.In what ww do the resources of a school and community determine how a school district will address the problem of talent development? 5.What constitutes a healthy respect for measurement error? Why is it so important for teachers to understand? 6.How is gifted education related to the more general notion of school improvement?

WHAT IS THE STUDY OF GIFTED STUDENTS?

This book is about what gifted students are like. It is not about how to teach gifted students, although the two topics have a good deal in common. As a result of our work in gifted education, the authors came to realize that the topic of what gifted students were like was not well understood. Whereas nearly everyone believes that gifted students are concerned with theoretical and aesthetic matters, no text seemed to pull together in one volume the empirica/ literature on such students. Since the 1960s, that literature has grown substantially, and we felt it was time to attempt a synthesis of this literature. Furthermore, it seemed to us that adults who were preparing to teach gifted students were the ones most in need of knowledge about the research concerning the characteristics of gifted students. Hence, we wrote this book for teachers who are now or will soon be working with such students. We hope it prompts teachers to arrive at new insights, and we hope it challenges widely held misconceptions effectively. This first chapter is an introduction to the issues and methods de-

1 I The Study cf Giftsd Students 3 scribed in the rest of the text. As you can tell from the focusing questions, it ties several themes together. It begins with a discussion of different types of defmitions and concludes by noting the connection between the improvement of schools and the improvement of programs that address the needs of very talented students. It covers a lot of ground. Our synthesis of what is known about teaching gifted students is briefly summarized in the last chapter of this text. in this text our main point is that instruction should have a lot to do with the kinds of students served in a particular program. We believe, therefore, that a text whose intention is to consider characteristics would be incomplete without a brief introduction to methods. For this reason, we not only include a chapter that provides an overview of issues related to instructional methods but also conclude many chapters with a section that discusses the instructional ramifications of the type of giftedness considered in the chapter.

DEFINITIONS

Definitions of giftedness reflect two perspectives: the theoretical and the functional. Each of these perspectives serves different purposes. Defini- tions of giftedness that are interested only in theory usually guide research or innovative programming. Functional definitions areusually designed for funding and regulating gifted education programs in the schools.

Theoretical Views of Giftedness Theoretical definitions ascribe giftedness to individuals who possess two attributes: (1) a particular kind of ablity and (2) ahigh level of that ability. The first artribute has been a greater source of controversy than the second, and it depends on the theoretician's perspective. Some psychologists (e.g.. Terman, 1925) have argued for a single, broad ability, such as "intelligence." Some (e.g., Gowan, 1978) have argued that "creativity" is more important. Many psychologists (e.g.,Guilford, 1967; Gardner, 1983) believe that there are a number of important mental abilities that exist independent of one another. These psychologists would include many dif - ferent types of abilities in their definitions of giftedness.

Social attributes.Paul Witty (1958) recognizes the social importance of giftedness when he defines giftedness as remarkable performance in any potentially valuable human endeavor. Tannenbaum (1983) discusses types of giftedness that meet society's explicit, implicit, and latent needs. They are, respectively, skills that are importantin meeting current manpower shortages, skills that are important for newly emerging specializations, and skills that are related to cultural enrichment, such as artistic skills. 2 4 Mt Shay 'falba Shu hods Another way of looking at giftedness as a social attribute, however, involves interpreting giftedness as a status rank (Goodnow, 1984). Accord- fr ; to this view, certain individuals are invested by others with the charac- usisticx of giftedness. They are thought of as gifted and treated as though they were gifted. From this perspective, giftednexs is not a characteristic of the individual at all. Instead, it is a trait ascribedrather arbitrarilyto the individual by a social group. For example, students who dress better than other 4tudents are perceived by their teachers to be brighter (Good & Brophy, I9e 7).

Psychological constructs.Some theoretical views c ...ail models of psychological reality, or "constructs." The work of several theoreticians involves such constructs. Galion (1869/1962), in Hereditary Genius, described intellectual formance in terms of the concept of deviation from the mean. He included many types of ability in his notion of giftedness:musical ability, athletic ability, and writing ability, among others. According to Gahon's view, however, ability alone was not sufficient forgenius to emerge. He maintained that "genius" was the interaction of ability, zeal, and stamina. Gahon believed that mental traits were inherited, and given the three necessary ingredients, genius would emerge no matter what the environmental conditions. Galton would not have used these three elements to locate gifted school children, however. He proposed no practical applications based upon his description of genius. ForGallon, the issue of genius had more to do with politics, economics, and empire than with schooling. In fact. the idea that schools could work to develop a high level of talent in students film many backgrounds would probably have struck Galtonlike many others of his eraas quite foolish. A recent definition, based on the work of Gakon and on more recent research on adult success, includes two of Ga hon's standards. Renzulli's (1978) definition of giftedness requires above average ability,creativity. and task commitment: characteristics of adults who make significant contributions to society. Unlike Gallon's definition, Renzulli's is intended to guide identification of gifted children. However, because it is based on adult characteristics, its applicability to school children is questionable. Children who show the combined characteristics of high ability, high creativity,and task commitment may indeed be talented, but the definitionoverlooks many children with uncommonpotential for academic achievement. Robert Sternberg's (1981) perspective on intelligence has also been applied to the problem of defining giftedness. His approach, based on the construct of information processing, places apremium on the di:Limey of cognitive processes. rather than on their rarity. Sternberg (1982) is particu-

1 3 The Study of Gifted Students 5 larly concerned with the individual's ability to handleunfamiliar (i.e., "non- entrenched") tasks and concepts. Terznan's (1925) definition of giftedness has boththeoretical and functional ramifications. He applied Galton's notionof the cotiparative rarity of talent to the study of highly intelligentschool children, and developed an instrument capable of distinguishingcomparative rarity. In his definition, giftedness is "the top 1% level in generalintellectual ability, as measured by the Stanford-Binet Intelligence Scale or acomparable instrument" (Terman, 1925, p. 43). Both Term in'sdefinition and his instrument have been very influential in psychologicalresearch and in school practice. The focus of Terman's work, however, like thatof Galton, was to describe talent and to draw conclusions about the utilityof gifted children to society. The description of talent is the aspect ofTerman's work that makes it theoretical. His interest in the utility of talent is the aspectthat makes it functional.

FUNCTIONAL VIEWS OF GIFTEDNESS

Functional views of giftedness apply more directly tothe practical issues of schooling. Hence, they tend to embody culturalvalues more explicitly than the theoretical views discussed previously.For example, functional definitions often serve as policy statements tohelp secure funding and provide a basis for regulations. Sometimes, however,functional definitions that accomplish the goals of society are not statedexplicitly. These functional views represent implicit, common-senseideas about talent. Such implicit ideas are strongly influenced by popularviews of the economic and political forces that shape the schoolingof children and the work of adults.

Pedagogical Definitions of Giftedness Pedagogical definitions specify thepopulation of childrer who are legitimately served through special programsfunded by the school system. These definitions must agree withthe philosophy of the school system and with the views of school administratorsand the community. Pedagogical definitions typically emphasize themission of the school to develop the potential of each individual student. Students'needs figure prominently in these definitions.

Federal definition.In the 1970s the U.S. governmentprovided two definitions of giftedness. These federaldefinitions of g:ftedness illustrate the pragmatic nature of institutionaldefinitions in two ways: (1) they have been constructed to suit political realitiesand (2) they are not theoretically

14 6 Ms Study of Gifted students coherent. In other words, such definitions are creations of the moment; they can be revised periodically, and often are. The 1972 definition of giftedness (Marland, 1972, p, 2) reads as follows;

Gifted and talented children are those identified by professionally qualified per:ions who, by virtue of outstanding abilities, are capable of high performance. These are children who require differentiated educational programs in order to realize their contribution to self and society. Children capable of high performance include those with demonstrated achievement and/or potential abifity in any of the following areas, Fing!y or in combination: 1.General intellectual ability, 2.Specific academic aptitude 3.Creative or productive thinking, 4.Leadership ability, 5.Visual and performing arts, 6.Psychomotor ability.

The 1978 federal definition eliminated "psychomotor ability" from among the areas of giftedness. The rrason for this exclusion was economic; experience showed that funds earmarked for gifted programs could be used for athletic programs. Sinc.! there was already a great deal of money available for athletic programs, and because the other kinds of giftedness in this functional definition were relatively neglected, "psychomotor ability" was removed from the definition. The recent definition, given in Section 902 of PL 95-561, The Gifted and Talented Children's Education Act of 1978, is worded slightly differently, but it includes the other five areas that are found in the 1972 definition:

The term "gifted and talented children" means children and, whenever applicable, youth who are identified at the preschool, elementary, or secondary level as possessing demonstrated or potential abilities that give evidence of high performance capabilities in areas such as intellectual, creative, specific academic, or leadership ability, or in the performing and visual arts, and who by reason thereof. require services or activities not ordinarily provided by the school.

The 1972 federal defmition has been criticized for its internal incon- sistency (e.g., Renzulli, 1978); and the 1978 definition has the same problems. The different types of ability listed in the definition include three hypothetical, content-free ability areas: intellectual ability, creative ability, and leadership ability. The two other types of ability are associated with disciplines: specific academic ability, and ability in the performing and visual arts.

1 5 The Study of Gifted Students 7

The federal definition has al.io been criticized byRenzulli (1978) for the lack of direction it provides local school personnel.He contends that the defmition is often misinterpreted. Some schooldistricts treat the six areas as mutually exclusive and set updifferent identification procedures for each. By contrast, other districts require only a high IQ score,regardless of the type of ability listed in their definition. Renzulli'scriticism calls into question both the theoretical coherence of the definition andits functional utility. Nevertheless, many states have adopteddefinitions like the federal definition (Council of State Directors of Programs for theGifted, 1985). Because so many states have adopted definitionssimilar to the federal definitim, we have chosen to organize the characteristicssections of this book on the basis of that definition. The chapters onintelligence, academic aptitude, creativity, leadership, and ability in the visimland performing arts elaborate the conceptsidentified with each of these kinds of talent and the characteristics associated with individuals whodemonstrate superiorizy in them.

Slate definilions.Major pragmatic considerations for school systems and other agencies serving gifted children are (1)priorities regarding the type or types of ability the systemis most concerned with developing and (2) budgetary constraints. These considerationsdetermine both the content of the definition and theeligibility criteria that limit the scope of the definition. State and local definitions of giftednessdetermine who will receive special instruction. Different definitions suggestdifferent identification procedures that locate different, if overlapping,populations of children. One school system may define giftedness asexceptional academic ability and give scope to its definition by requiring scoresabove the 95th percentile on standardized individually administeredachievement or aptitude tests. Another system, with different educationalpriorities, may define giftedness as exceptional ability in any ofseveral areas, including, for example, academics, performing arts, or visual arts.Given the same proportion of funds to spend on gifted edi,...ation, theschool district with the broader definition could limit its gifted populaiion byestablishing a higher criterion level in each area. In general, the morelimited the funds, the higher the level of giftedness required for eligibility forspecial programs. Well over half the states' definitions include mostof the abiLies included in the federal definition. Forty-four states includeintellectual ability, 42 specific academic ability, 36 creative ability. 32ability in the arts, and 23 leadership ability (Council of State Directors ofPrograms for the Gifted, 1985). The highest priority is intellectualability, perhaps reflecting the long history of research interest in that area. Of those statesthat include an

1 fi 8 The Study of Gifted Studeuts

142 cutoff in their definition, the most commonly named level is two standard deviations above the mean, or about the 97th or 98th percentile (Council of State Directors of Programs for the Gifted, 1985). Marland (1972) estimated that in applying a broad definition of giftedness, such as that of the federal definition, at leo( 3 to 5 percent of the total school-age population would considered gifted. It may be that states with mandates that require special education for all identified gifted children tend toward narrower definitions, such as the high-IQ definition, because they cannot afford to serve 5 percent of the school-age population in gifted programs. Where there is no mandate, and where local school systems determine whether or not to serve gifted children and how many and what kind to serve, a broad definition is more likely to be adopted, even though it will probably not be implemented very well (cf. Mitchell, 1981).

Instrumental Definitions Two rationales support the provision of special programs for gifted students. The first is concern for the development of individual potential, a concern that strongly influences pedagogical definitions of giftedness. The second rationale is development of talent as a national resource. The United States is not alone in expressing its concern for national security through efforts to find and use the talents of gifted individuals; all societies choose to nurture gifted individuals in accordance with the society's perceived values and needs (Kitano & Kirby, 1986). Indeed, concern over the interests of business and national security help to turn away some of the suspicions that gifted education arouses in citizens. Without such a compelling rationale, it is unlikely that the public would allow the expenditure of any funds for gifted programs.

National security.According to Laycock, society must have the contributions of gifted individuals. It needs them for survival and for cultural enrichment. Laycock's book, Gifted Children (1979, preface), begins, "Gifted children are any society's prime asset." This sentiment, held by many citizens, offsets their resistance to giftededucation. This feeling is confounded, however, with the suspicion that gifzed children are not only an asset to society, but also threatening to the status quo. Suchfears seemed justified during the late 1960s: The leadership of the student protest movement was composed largely of uncommonlytalented (and uncommonly advantaged) students. Gardner (1961) explains society's mixed feelings toward excellence. Such feelings probably condition the public's reactions toward gifted individuals. According to Gardler, the critical lines of tension in our society are between the emphasis individual performance and the restraints on 7 The Study of Wird Students 9

individual performance. f hese conflicting interests help to create what Kirk and Gallagher (1983) have called the love-hate relationship that exists between American society and its gifted individuals. During times of political crisis, attention to the development of talent increases and consequently so does programming for the gifted. Tannen- baum (1981) has identified two peaks of interest in gifted children: the five years following the 1957 launching of Sputnik and the five years following the 1975 withdrawal of U.S. troops from Vietnam. Between those times, there was little interest in the gifted. In 1983, A Nation At Risk, published by the National Commission on Excellence in Education, reported troubling findings: U.S. students seemed to compare unfavorably with students from other world powers. The report emphasized the need for more attention to academic subjects, and it recommended that the federal government help fund and support efforts to meet the needs of the gifted and talented.

Business definitions.Economic supremacy is considered by many to be essential to national security, and business has had a determining influence on education in the United States. This influence is documented by Raymond Callahan (1962) in Education and the Cult of Efficiency. According to Callahan, the influence of business over education has grown steadily since the early twentieth century. One of the first indications of business influence was change in the composition of school boards. They changed from large. mixed-interest groups to small grown of businessmen. This change, it was hoped, would enable schools to be run more like businesses, efficiently and economically. At the same time, there was an effort to make the curriculum of the schools more practical. Vocational schools and courses were established, classical studies declined, and a strong current of anti-intellectualism began to appear, evidenced by such phrases as "mere scholastic education" or "mere book learning" (Callahan, 1962, p. 8). Andrew Carnegie was one of the leading critics of the traditional curriculum and one of the strongest proponents of practical education. In his view, the secret of his own success and that of other business leaders was not book learning. He saw more value in ingenuity and common sense. Carnegie's view and that of other businessmen put pressure on school superintendents to develop practical curricula. Historical records confirm that the superintendents responded to the pressures by making changes in the schools. Callahan (1962) quotes proceedings from the National Education Association that document such changes. For example, he quotes a bank president's congratulatory address to a group of school superintendents for their contribution to America's material progress. The bank president maintained that such progress was "a result of getting away, to an extent, from the mere scholastic education, and developing the practical side, making the school the place to learn how

1 10 The Study al Wird Shields to manufacture,." (NEA Proceedings. 1901, cited in Callahan, 1962, P. 9). Given the businessmen's interest in ingenuity and industry, business perspectives on giftedness have focused more on creativity than on intelligence. Divergent and productive thinking are important commoditiesin the business world, which has to keep marketing novel products and developing more efficient ways to solve problems. Programs for increasing creativity have emphasized industrial and technological applications. Osborn (1953), for example, devised a number of techniques, such as brainstorming, to help advertising personnel solve problems. Such programs and techniques have been imported directly from business to gifted education.

CHARACTERISTICS: EMPIRICAL VIEWS OF GIFTEDNESS

As the preceding discussion indicated, there is little consensus about what constitutes giftedness. Diverse interests view giftedness in rather different ways. How can one possibly know which view of giftedness is correct? Surely, theoreticians and practitioners can debate the matter indefinitely. Can a perpetual debate be of much use to educators who want to help talented children get the most out of their years in public schools? The answer to that question is a qualified yes. Over the long term, knowledge is advanced by such debates in c.cry field. On the other hand, the ambigu- ities of that debate complicate the educator's immediate job. The debate offers little in the way of practical guidance for designing good school programs or for developing further the ones already in existence. One way to learn more about the concept of giftedness as it pertains to children in schools is to review the empirical research on gifted children. This section of the chapter begins by discussing the various kinds of empirical research about the characteristics of able students. Then it discusses the moreifficult question of interpreting empirical results.

Research About Students' Characteristics Studies of the characteristics of individuals who are said to be gifted can be classified into three groups: (1) case studies, (2)quantitative cross- sectional studies, and (3) longitudinal studies. Such studies will be reviewed throughout subsequent chapters, so a brief introduction to these general types of study follows. Readers should remember that empirical research cannot ultimately determine which theoretical view of giftedness is best in general. it can help thoughtful readers decide the question for themselves, and it can help them make better-informed practical decisions with respect to children in schools.

1 9 Ms Stay of Gilled Students 11

Case studies.Case studies describe particular individuals or a number of individuals in depth. An empirical case study stays very close to verifiable observations (there are many other types of case studies). Hence, such a study typically includes a wealth of factual data and comparatively little critical analysis. Most case studies make rather interesting reading. Reading a number of case studies, however, ultimately yields only an im- pressionistic view of a category of individua1s. Techniques other than the simple case study are needed if hypotheses about a category of individuals are to be examined carefully. Case studies are reported throughout this textbook. Most of them are anecdotal in nature, and a number of them were gathered before inferen- tial statistics became established in the social sciences. Perhaps the most famous American case studies in gifted education are those prepared by Leta Hollingworth (1942). Intrigued by the use of the Stanford-Binet Intelligence Scale, Hollingworth described in detail 12 children who scored very high (above 180 I(2) on the test. (Hollingworth had also done earlier work with very low-scoring children.) She reported a variety of data about these high-scoring children, including information about their birth and infancy, school progress, and later development. Two European scholars, Jean Piaget and Vad'im Krutetskii, have produced substantial research using what might be termed case study methods. Neither of these scholars, however, has used very sophisticated statistical techniques. Jean Piaget was a renowned Swiss psychologist. His unit of analysis was an idealized typical child, rather than actual exceptional individuals (Fish- er, 1987). Piaget's interest seems abstruse, remote, and impractical to many American educators, but it reflects his European intellectual background. He was interested more in philosophical and metaphysical questions (e.g., how is knowledge related to human growth?) than in practical pedagogical questions (e.g., what is good instruction in our schools?). Since the early 1970s American scholars, however, have adapted Piaget's techniques to the American context of research. Some of the work that has been influenced by the Piagetiap method is reported in Chapter 8. The most original use of case stucLes in research about exceptumal talent is probably a Soviet achievement. To investigate mathematical talent, the Soviet psychologist Vad'im Krutetskii used a method of case study with over 200 children. His work, which began in the early 1960s, became available in the United States in the mid-1970s, largely through the efforts of Irving Wirszup and his colleagues at the University of Chicago. American researchers have just begun to exploit the methods developed by Krii- tetskii. Case studie are used in enterprises other than basic research. Some- times they are used to present the results of research in an anecdotal fashion. The Study of Mathematically Precocious Youth, for example, has

4.10 12 The Study of Geed Students often published case studies to dramatize the accomplishments of the talented students it serves. Case studies are also sometimes recommended as a means to develop program plans forgifted students (e.g.. Baldwin, 1978; Maker, 1982). Studies of the sort conducted by Krutetskii and Piaget are very expensive. They require intense interaction with each subject and equallyintense analysis of the data accumulated during the interactions. Oftenthe work takes decades to complete; in a sense, the scope of case study projects like these is so great that the work can never be considered complete. Case studies are, in the United States, typically reserved forinstances in which it is difficult to obtain a sufficiently large sample topermit quantitative analyses. In the United States, studies typically employresearch designs that involve many subjects, a short time frame, and produceresults at substantially less cost persubject than the European case study research cited previously. American empirical studies usually opt for efficiency of data collection and analysis, whereas European studiesoften opt for a complex theoretical analysis of data. Case studiesoffer more opportunity for such analysis than thequantitativemethods favored by empirical researchers in the United States.

Quantitative cross-sectional studies.Most of the studies reviewed in this textbook are based on quantitative analyses of theperformance of selected samples on a variety of formal tests. Of thesestudies, most aretram sectionsof the performance of the individuals in the sample. Quantitative methods are based on the statistical notionof generalizability. A number of conditionsmost of which can be measured numericallymust be met before obtained findings can be said to represent a general finding. Two ofthese conditions are especially important. First, the sample must be representative of the larger group forwhich it will be claimed that the findings arecharacteri.stic.Second, the sample must be large enough so that theresearchers can say the findings are not due to chance. This second condition means that if thtresearcher is studying a large number of characteristics, or looking at smalldifferences in characteristics, the sample must be rather large, perhapsincluding several hundred or several thousand subjects. These conditions arevirtually impossible to meet in typical case study methods. The cross-sectional studies reported in subsequentchapters are of several sorts. Among these are correlational studies. Manycorrelational studies were conducted in the earlier part of the century,following the development of the technique by Karl Pearson. A "Pearson product-moment" is the most common typeof correlation. It measures the strength of one variable'srelationship to another, on the assumption that the relationship can bedisplayed graphically as a straight line. The more closely the relationshipresembles a straight line, 2 1 The Stody of Gifted Students 13 the stronger the correlation. High correlations between two conditions (e.g., performance on an IQ test and success in school) can be interpreted as indicating a characteristic relationship. The results of correlational studies are reported throughout this textbook. For example, the relationship between IQ and early reading is a topic in both Chapter 3 and Chapter 6; the relationship between academic performance and professional performance is a topic in Chapter 5. The patterns among correlations of related characteristics have also intrigued quantitative researchers working with cross-sectional data. These studies are of basically two types. One method, developed very early in the twentieth century, is called "factor analysis" and has been applied widely to the study of individual characteristics. The goal of factor analysis is to create a kind of synthetic characteristic from a number of correlations among real characteristics. Early researchers who used factor analysis gave many tests to a group of subjects and correlated all the subjects' performances on each pair of tests. Then they could construct a "factor" (a synthetic measure) withwhich most of the tests correlated highly. In this textbookfactor analysis figures most strongly in discussions about creativity, intelligence, and academic talent (Chapters 3, 4, and 6). Stephen Gould has developed a very understandable explanation of factor analysis (Gould, 1981, pp. 243-250). Another method of examining patterns of correlation is called "multiple regression." This method is used when researchers want to assign the sources of influence in one measure (calledthe "dependent variable") to a number of other measures (called "independent variables"). This technique has been used less frequently to study psychological and educational characteristics than to study the effectiveness of instruction or to examine the relationships among sociological characteristics. Some works that use multiple regression, however, are referenced in Chapters 2 and 11.

Longitudinal studies.Longitudinal studies follow a sample (called a "cohort") for a long period of time in order to chart their development. Lewis Terman identified such a cohort in the early 1920s, and he and his associates followed the progress of the cohort throughout their years in public school, in college and graduate school, in their professions, and into retirement. The Terman study is the most extensive longitudinal study in educational research; and its topic is, of course, gifted students. The study illustrates very well the difficulties of conductinglongitudinal research of this scope. For example, standards ofresearch changed dramatically between 1920 and 1980. Terman would have designed a much different study,if, in 1920, he had had access to the tools and techniques (including computers) that were available even so long ago as 1960. Nonetheless, the Terman study has collected masses of data that oth-

22 14 7* s Study of Gifted Students er researchers have begun to reanalyze. Some of this research isreported in Chapters 3, 10, and I I. In general, however, longitudinal studies are rare not only because of the technical problems mentioned previously,but also because the logistica/ problems of staff continuity and continual funding are difficult to resolve.

INTERPRETING EMPIRICAL RESULTS

Readers should remember that the basic material used in this textbook (American quantitative research) strongly influences the presentation. The empirical view presented here is a peculiarly American view. We believe that fact makes this research particularly applicable to the schooling of very talented students in the United States. American researchers are often concerned with testing carefully defined hypotheses that seem to have clear implications for practice. The way in which hypotheses are defined and the way in which results are derived aLso convey the impression of practicality. Hypotheses are designed to be tested statistically so that they can be shown to be either true or false within specified limits. Hence, both the hypotheses and the results of study can be stated simply. These facts help lend the status of objectivity to obtained results. There is another sense in which quantitative research methods can be said to be practical. They make possible the use of more subjects at lower cost in a shorter time period.

Recognizing Assumptions Empirical study carries with it assumptions that those who use it must understand. Too often, educat are unaware of these assumptions. For a variety of reasons they may be unaware of the intellectual traditions that shape empirical social science research. Sometimes, teacher preparation programs include little theoretical work;sometimes introductory coUrSes include no perspective other than that of empirical social science. Here we want merely to point out some of the major assumptions of most American empirical science and to alert the reader tothe existence of more complex views of education in general. Theseassumptions entail (1) a focus on the individual, (2) faith in the measurability of behavior as well as of invisible traits and constructs, and (3) faith in the utility of research as a guide to practice.

Focus on the individual.In a pedagogical specialty like gifted education, virtually all the empirical research focuses on individuals. Teachers and other educators want to know what particular children who are said to be "gifted" are like; and they want to know, on the basis of such findings, 23 ittgstudy etivdMandl! 15 how such children should be treated in schools. Hence, in giftededucation, it seems especially natural that individuals should be the focus of research. The values of our culture stress the importance of the individual, and they tend to deny the importance of other entities, such as social class, that might serve as "units of analysis." It is important to remember this point when one interprets the empirical view of giftedness.

Faith in measurability.Most of the studies reported in this textbook are quantitative cross sections. They are often largestudies of the performances of many individuals on tests. The tests assumethat the behaviors, traits, or constructs they reflect are measurable. A critical requirement of such research is that the instruments used be valid and the measurements obtained be reliable. There are two main problems with measurability. First, there is an important technical problem. The requirements of validity and reliabity are not always met. (Some critics would saythey are seldom met.) Research instruments are often "technically inadequate." That is, they are not good enough to perform some of the tasks for which they are used. Second, there is an important theoretical problem. In defining hypotheses very narrowlyso that they can be tested statisticallyresearch- ers usually exclude from consideration muchthat might be relevant to the question. Intelligence tests are a good case in point. Alfred Binet, the real originator of intelligence tests, did not himself believe that aphenomenon so complex as intelligence couldbe measured easily. He certainly did not believe his test measured intelligence. It did, he believed, measurethe likelihood that a student would benefit from instruction in the regular classroom. The test had a narrow practical purpose. When scholarsand educators use IQ tests to define intelligence, they are straying very far from the original purpose. The very great controversy that has emerged aroundthis sort of misapplication of IQ tests is considered at length in Chapter 2. The discussion in that chapter is really a "case study" about the limitations ofempirical research in the social sciences.

Faith in research as a guide to practice.American educational researchers expect that their work will help shape what goes onin classrooms. They are, however, often frustrated in this expectation. Itis no wonder. Researchers study a wide assortment of problems, using a wide varietyof methods, and they often reach conclusions that are contradictory. Itis probably accurate to say that research influences what goes onin classrooms only indirectly (Buchmann,1987), Although every research study that appears should not guide practice, it is true that understanding the reading, thinking, andwriting that others have done about education can be a very good guide toworking with 16 The Stsuly of Gifted Students students in schools. Teachers must be active participants in this endeavor, however. They cannot be told what to do: They must read, think, and decide for themselves. Disciplined inquiry cannot yield a formula for operating the schools. Too many things change in the course of a decade; schools and children are too different from one another, and so are teachers. Hence in reading this textbook, educators should understand that we have tried to present empirical research about gifted students in a way that will help them think and decide for themselves what may make sense in the classroom. Chapter 12, however, presents our own interpretation of what makes sense in the classroom.

Applying Research About Characteristics Research about characteristics addresses the question of what gifted students are like. In general, the phenomenon of concern to educators interested in gifted children is uncommon (or "exceptional") talent. Many kinds of exceptional talent, and many circumstances under which exceptional talent exists, are discussed in subsequent chapters. Of what use is this knowledge? Since the report Sidney Marland made in 1972 about gifted students to the Congress, programs for gifted children have grown rapidly. The rate of increase in these programs seems to be slowing. A great deal needs to be done to improve these programs (Cox, Daniel, & Boston, 1985). Some programs are elitist (Weiler, 1978) or even anti-intellectual (A.Howley, 1986; C. Howley, 1986). The source of these problems, it seems to us, is a mismatch between the kinds of students identified (i.e., the characteristics of identified students) and the types of programs that are offered to students. This problem cripples the effectiveness of many programs. These programs do not supply the kind of instruction required by the students they serve. Across the nation, many states have adopted a definition of giftedness that includes many of the types of talent described in subsequent chapters of this textbook. At the same time, the programs they offer serve primarily high-1Q children (Council of State Directors of Programs for the Gifted, 1985); and the high-IQ children served in these programs do not have access to the flexible pacing most appropriate for them (Cox etal., 1985). If the schools are serious about serving a wide variety of talent general intellectual ability, specific academic talent, artistic talent, creative talent, leadership ability, and manual talent (e.g.. gifted students in vocational schools)then educators will need to know much more about students who are talented in each of those categories. This is a mammoth undertaking; it is as if educators concerned with mentally retarded children advocated five or six subcategories of mental retardation, with differentiated programming required for each sub-

25 The Study of Gifted Stade*: 17 category. In gifted education, this plan has not only been proposed. but it has been translated into state and local regulations. In applying empirical findings about creative individvals, artists, high-IQ students, early readers, and precocious math students, educators need to consider several problems simultaneously:

Which instruments select students who exhibit these characteristics? Which characteristics can be addressed by instruction? What kind of instruction is needed? How can instruction be arranged efficiently?

Chapter 12 deals with some of these questions, primarily for students who demonstrate general intellectual ability and specific academic talent. In general, however, we note that although other kinds of talented students can be described, it is impossible to locate technically adequate instruments to identify particular individuals who exhibit those characteristics. Despite this impediment, however, the empirical literature on the characteristics of gifted children suggests that many gifted children possess multiple talents. Perhaps the tendency to possess multiple talents is the reason multicategorical definitions of giftedness identify so many high-IQ students and so few students with specific talents. If this is the case, then most gifted programs, as Cox and colleagues (1985) note, also need to improve the comprehensiveness of their offerings.

TALENT IN THE SCHOOL SETTING

The development of talent seems to be a logical concern of schools. Most school districts state goals such as this: to enable each student to develop to his or her fullest potential. Such goals are ambitious in the context of universal schooling. Perhaps the difficulty of accomplishing these sorts of goals leads schools to identify only a small group of students who show the greatest promise. If schools view a minority of students as talented, then the development of talent becomes a more limited, and hence, more manageable enterprise. Global definitions of talent, however, suggest a different perspective. If schools cannot predict the adult contributions that will be made by children, then they should prepare all children to perform optimally. Com- mon sense suggests that a well-educated majority will collectively accomplish more than a well-educated minority. In spite of this logic, schools may be unable, in practice, to accomplish such an encompassing goal. Political, economic, and even pedagogical realities militate against this goal. This section of the chapter examines the schools' role in cultivating the potential of school children. It considers issues related to ( I) the identi-

C.0 p0 18 The Study of COM Studios* fication of talents, (2) the social context of talent development, and (3) the educational response to talent.

FIncHng Talent The method that a school distric! &nes to identify its talented students depends, in large measure, on its definition of talent. It also depends on the resources available to the school district, though this dependency is less obvious. Resources govern the quality and quantity of services available to those students who are considered to be talented. They also govern the size of the group that can be identified as talented. In wealthy, suburban school districts, for example, a large number of students are presumed to have the academic talents needed for the successful completion of undergraduate programs in selective colleges and universities. Such districts provide students with the rigorous educational ex- paiences that will prepare them for success in &lege. This sort of preparation depends on the district's ability to 1,, ;,.together a number of pedagogical resources: excellent teachers, a chatie..ging curriculum, supportive guidance counseling, and a community of receptive students. Un- common leadership can help other districts provide these things. On the other hand, poorer districts have in general a much more difficult task, and not only because they lack material resources. Such districts may not expect to send a majority of students to college. The expectation of these schools mirrors the expectation of the communities they serve. College attendance may not be a value of such communities; it may not even be perceived as a promising avenue for economic advancement. Such districts may identify only a small percentage of students as talented, or they may fail to identify any students as talented. Most school districts, however, are neither wealthy nor impoverished. Instead, they serve students from a variety of economic backgrounds. In these districts, the identification of talent often reflects, though not per- fectly, the social structure of the community: The children of local professionals are often placed in the school's gifted program. Less often are the children of unskilled workers selected for such programs. The following interpretation is one way to synthesize the preceding observations: The wealthier the district, the more is talent perceived as universal; and the poorer the district, the more is talent considered a rarity. Some hereditarians (e.g., Jensen, 1973) explain this educational disparity as a difference in the structure of reality. According to their argument, society rewards talent through its economic structures. Hence, wealth is a proximate measure of talent. The logic of the hereditarian position leads to the following conclusion: Wealthy school districts identify more talented students than poorer districts because such districts contain more .nted students.

2 7 The Study of Gifted Students 19

Psychometric definitions of talent have been used to supportthe hereditarian position. After all, high socioeconomic status(high-SES) students do score higher on intelligence tests thando low-SES students. A more generous view oftalent, however, accords environment amuch greater role. Such a view suggeststhat children's social environments condition their performance on standardizedintelligence tests and other measures of performance. Even moreimportant, this perspective emphasizes the role of schooling in the cultivationof talent.

Rarity and universality.Definitions of giftedness vary dramatically in their scope. According to some defmitions,giftedness represents excellence in any area of human endeavor. Otherdefinitions relate giftedness much more closely to academic knowledge.Both perspectives, however, have important ramifications for schools. Although schools might want to develop all ofthe talents of all their students, very few of them have the resources toaccomplish this goal. Given these conditions, schools may wisely opt toidentify as gifted the students who are least likely to benefit fromgrade-level instruction in the regular classroom. At the same time, however,schools can increase their responsiveness to the needs of all students. Theliterature on school effectiveness reiterates this position: Certainschool practices are capable of cultivating higher achievement, even inschools in the poorest communities. Significant changes in the priorities of our society are necessarybe- for- schools can attempt to develop alltalents. Universal schooling has iimitceffectsimparting basic skills and propagating theknowledge be- liev -cl necessary for good citizenship. Inschools designed to accomplish these minimal outcomes, it can be difficuh to nurture somekinds of uncommon talent. Inspite of their limited resources, however,public schools min address certain typesof uncommon talent. In particular, thedevelopment of aradrnsic talentdoes seem possible in our educational system as presently structured.

Practical method, of identification.When schools choose to identify academically talented students, they have severalalternatives. They may base their identification on students' grades,intelligence test scores, or achievement test scores. Many districts attempt to identify studentswho are talented in areas other than academics. For example, talentin sports is an area addressed by most public schools.School athletic programs provide an interestingmodel of identification practice. Large groupsof students are provided with training in basic skills through the school'sphysical education programs, and those who perform well or who enjoythe sports activities choose to try out for teams. Usually the district supports ahierarchy of teams: Intramural teams are the least selective,whereas varsity teams are the most selective.

2S 20 The Study of Gifted Students

Students who are selected for teams are drilled in the skills and strategies of the sport. From among those who are trained, the most excellent are selected to perform in competitive events. Few other school endeavors give so much opportunity for talent development. In schools that support excellent drama, dance, music, and art programs, however, the selection process resembles the one for athletics. Talented students audition for parts in plays or for places in the orchestra. Art students prepare portfolios of their best work. In such programs, an expert in art, drama, dance, or music, or panel of such experts, selects the students with the most potential. Some districts use a relatively broad definition of giftedness to select students to participate in gifted programs. They may use multiple selection criteria in order to restria the number of students who are identified or they may use multiple selection criteria in order to incrrase the number of students identified. In either case, however, the identified group will be composed of students with a variety of abilities and needs. Students identified on the basis of one tyr_ of measure vary considerably from those identified with another type of measure. Because these students do not have similar educational needs, they cannot be served well in a single gifted program; yet rarely do districts maintain separate gifted programs for intellectually gifted students, creatively gifted students, and students with leadership potential.

Error and equity.Educators pay more or less attention to issues of equity depending on their purposes for identifying talented student.s (e.g., development of individual potential or a concern for the national security). Ensuring equity complicates fulfillment of these purposes. In order to ensure equity, educators must developidentification procedures that account for social inequities. Such procedures can be perceived asimpractical. Some educators recommend using subjective measures of talent in order to offset the cultural bias of standardized tests (cf. Council of State Directors of Programs for the Gifted, 1985). Districts that use subjective measures reason that, if they avoid using standardized testsaltogether, they will be free to include more poor and minority students in programs for the gifted. Although this observation is correct, it means that identification depends solely on the good will of those who administer the program. This situation forces districts to rely on unreliable measures that may have little relationship to talent. There is no reason to believe that subjective measures of leadership ability, creativity, or accomplishment will identify the most talented students from among poor, black, or Hispanic families. D.!pending on the good will of the subjective evaluators, districts are fr:e to include more poor students in gifted programs; however, they are also free toexclude all

29 Tim Study Gifted Students 21 poor students when they usesubjective measures. This degree of uncertainty is inherent in the practice of subjective measurement. Given what is known about cultural and racial prejudice,it is probably very unwise to put a greatdeal of faith in the fairness of subjective measures, or the good will of programadministrators. Objective measures (e.g., standardized tests) have the advantage of allowingeducators to identify and even quantify their degree of prejudice. Weknow, for example, what the differences are between the black mean and thewhite mean on IQ tests. We are very unlikely,by contrast, to understand how or to what degree unexamined assumptions influence subjectivejudgments. Critical to the use of objective measures, however, is ahealthy regard for error. Educators should remember that anyobtained score always reflects some degree of error (see Box 1-1). Once weunderstand the degree of test ermr that is reflected in the scores of arelatively heterogeneous forming sample, we can begin to appreciate thedegree of error that might be reflected in the comparison of a minoritystudent's scores to those of a predominantly white, middle-class forming group.This understanding leads to the recommendation that minoritystudents be compared to a group of their peers.Statistical methods for making such comparisons exist. The process is sometimes referred to aslocal forming.

BOX 1-1 How True Is tv., Obtained Score?

What does it inean when a student obtains aparticular 5031T on an IQ test? Is a score of 126 significantly different from a score of 130?These questions can be answered through an explanation of the related concepts of test"reliability" and "standard error of measure." The reliability of a measure is the degree towhich successive scores obtained on an infinite number of potential administrations of a test clusteraround a hypothetical "true" score. Statistically, test reliabilityis reported as a correlation. The dispersionof scores around the hypothetical true score can becalculated as a standard deviation. This special kind of standard deviation is calledthe "standard error of measure' (SEM). Using the standard error of measure, evaluators candetermine the likelihood thatastudent's obtainedscore reflects a true score within aparticular range. Typically, test scores obtained on individually administeredIQ tests are reported on a standard score scale (mean = 100, S = 15 or S = 16,depending on the test). One SEM on such a test may be 3, 4, or 5points, again, depending on the test. What this means about a given obtained scoreis that it is likely to fall within a range that is equal to twice the size of the SEM.This range is known as the "confidence interval."If the SEM is 5 points, then the likely range is 10points. This calculation reflects the fact that we speak of a score in consideration ofthe test's SEM. Usually scores are reported as the obtained scoreplusor minus 1 SEM. It ispossible, however, toreport a score in considerationof 2, or even 3, SEMs. The more error is taken into consideration in the reporting of the score,the more likely will the predicted range be accurate. 3 n The Study ef Gifird Shassu

What does a score mean when it is reported in consideration of 1 SEM? It means that in approximately 68 percort ofthehypothetical administrations of the test, the obtained score will fall within the ramp described by the SEM (i.e., the confidence interval). If, for example, a student obtains a score of 130 on a Stanford-Binet IQ test, this obtained score should be expressed in terms of the confidence interval. One SEM represents a 68 percent confidence interval. Hence, the student's score will probably fall within the range of 125 to 135. How sure can we be that the score will fall within this range? We -an be 68 percent sure. What can we do, if we want to be more certain of a score range? We can increase the confidence interval by reporting a score in terms of more than 1 SEM. With a range of 2 SEMs we can be 95 percent confident; with a range of 3 SEMs we can be 99 percent confident Returning to our example, we find thatthe obtainedIQ score of 130 stands a 95 percent chance of representing a °true score that falls in the range of 120 to 140 and a 99 percent chance of representing a alrue score in the range of 115 to 145. in other words, we can be quiteceitain that the student's intellectualability falls somewhere in the above average to very superior range. This degree of certainty doesn't help us decide, however, whether orat to place the student in the gifted program. In fact, it makes our decision more difficult. How can we be sure that this student istrulygifted? The truth is: We can't.

Rewarding and Punishing Talent Once aschool district identifies talented students, by whatever means, it must decide what to do for these students. Almost always identification is followed by placement into some type of program. The nature ofthe program, and how it fits into the school curriculum as a whole, depends on the way talent is viewed in the school and the community. Scholars, including both educitors and historians, differ in their interpretations of how talent is treated in the United States. According to some scholars, talent is rewarded through the economic and political structures of society. According to others, it is not rewarded at all, but is even punished. The view that talent leads to disproportionate rewards suggests that programs for developing talent are likely to support elitist aims (Feld- man, 1979). The view that talent is not appreciated, on theother hand, suggests that programs for developing talent exist in an unsupportive and even anti-intellectualcontext. Elitism. Todetermine whether or not gifted programs are elitist, we need to analyze their aims. Programs that do make extensive provisions to cultivate talents are probably not elitist. They serve a well-defined group of students, their purposes are consonant with the general aims of education, and they provide instruction in response to the identification of talents. When such programs make a special effort to include poor and minority students, they probably counter elitism.

31 The Shod, el Gifted &acids 23

Other programs, however, offer talented students special oppor- olnities that do not contribute in substantial ways to thedevelopment of their talents. These programs provide "enriching" experiences such as field trips and brainstorming sessions. Educators, however, would have a difficult time demonstrating that such programs are not elitist. Since all students would benefit from the activities included in such programs,why should only a few be permitted to participate? (For a thoughtful discussion of the elitist nature of such programs, see Weiler's 1978 article, "TheAlpha Children: California's Brave New World for the Gifted." For a morethe- oretical discussion, see Aimee Howley's 1986 article, "GiftedEducation and the Spectre of Elitism.") Anti-inteliedssalisns. A number of scholars have observed the anti- intellectual character of schools in the United States. Coleman's(1961) study, Adolescent Society, suggests that high schools reward prowessin sports and physical appearance much more than they rewardscholarship. Similar impressions emerge from the naturalistic studies of working-class and middle-class schools (e.g., Wilcox, 1982). By contraft, studies ofclassrooms in exclusive private schools reveal an emphasis on scholarshipand critical thinking (e.g., Anyon, 1980; Jackson, 1981). In order for talent development to be pursued systematicallyin U.S. school districts, communities will need to support intellectualvalues. Such values seem, however, to conflict with the practical values ofthe U.S. political economy. Until intellectual and artistic work isendorsed for its own sake, rather than for its utility, programs for talentedstudents will be in jeopardy. Such programs will, at the whim of legislators, schoolboards, or school administrators, be replaced by other programsthatat a given timeappear to be of more practical value.

Developing Talent In spite of claims that it might be elitist and in spite ofthe sometimes anti-intellectual context of schooling, talent developmentnevertheless seems a worthy enterprise forpublic schools to undertake. It is important work for several reasons;

It enables individuals to achieve to their greatest potential. It prevents students from encountering the wasteful boredomof work that is not challenging. It trains even the most facile students in the ethos of hard work. It advances the intellectual and aesthetic legacy of Westerncivilization.

Schools that choose to engage in talent development can(I) implement programs that meet the needsof a small minority of students who are

32 24 Tkr StudyofGifted Studasts identified as talented and (2) improve instructional programs for all students. These two approaches need not be mutually exclusive, so long as gifted education does not provide programs that are elitist.

Gift! d programs.Gifted programs should be based on the characteristics of the students who are identified as talented, in specific, and the characteristics of schools, in general. Most manageable within public schools are programs that address the needs of academically talented students. In general, the business of public schools is to cultivate academic learning. Hence, it would be practicable for schools to identify the most academically apt students and provide them with advanced instruction. This practice would meet the needs of such students, and it would enhance the academic climate of schools. Even when the content of public school gifted programs is academic, the instructional methods of the program may not conform to the characteristics of academically talented students. Too often advanced content is not taught at an advanced conceptual level. Since academically talented students are usually characterized by superior abstract reasoning, gifted programs should provide learning experiences that are conceptually complex. Such experiences are based on extensive involvement with textual materials. They provide the opportunity to critique, debate, discuss, and apply ideas.

School improvement.Talent development is best accomplished in schools that use effective practices. In such schools, the need to identify a select group of the most capable students may be unnecessary. Such schools provide a climate that supports academic learning. They enable all students to progress rapidly through instruction at their level of performance, and they encourage poor and minority students to excel academically. Even in schools where appropriate practices have a strong positive effect on academic achievement, talent development may be limited. Such schools may deliver high-quality instruction in basic skills but may neglect to provide experiences that encourage critical thinking or foster respect for intellectual endeavors. According to Timar and Kirp (1987):

Excellence is a difficult product for educational policymakers to deliver. High academic standards may require some individualsstudents, teachers, parents, administratorsto change their attitudes about schooling. One of the underlying assumptions, for example, is that students can be made to value education intrinsically. An appreciation of education for its own sake, it is thought, can be fostered with some clever policy engineering. Making class periods longer, requiring more school days each year, promoting enrollment in academic courses, creating a curriculum that emphasizes basics, and cur- tailing students' extracurricular activities are generally thought to be ways of redressing mediocrity. (p. 312) ,

The Stay tGft.i Students 25 Schoolseven the most effectivemay have a difficulttime providing appropriate programs for developing certain kinds oftalent. Because of limited resources, schools often cannot provide theextensive sort of training in art, music, dance, or drama that is requiredfor the d Telop- ment of talents in these areas. Talentdevelopment in music, for example, requires quite extraordinary resources; private lessons,opportunities for solo performance, opportunities to perform inensembles, ear training, instruction in music history, and instruction in musictheory. Such a range of educational experiences is available in only alimited number of public schools, usually so-called magnet schools in largeurban areas. When schools are unable to provide programs forcertain sorts (A. talent development, they should encouragecapable students to seek appru- plate cut-of-school training. Schools with an abiding .st in providing poor and minority studentswith equal access to programseven those that are offered outside ofthe schoolshould help such students obtainthe necessary scholarship assistance.

Summary This chapter introduced the notion ofcharacteristics in gifted education. The discussion examined definitions ofgiftednesstheoretical views of giftedness as a social attribute or as a psychological construct,and functional views of giftedness for educational purposesand for other specific goals. The discussion showed that functionaldefinitions had clearer implications for practice, whereas theoretical definitionsoffered wider scope for reflection about human nature, social interaction,and talent development throughort society in general. While definitions provide a context forunderstanding giftedness, they cannot give a detailed view of whatgifted individuals are like. Defini- tions, after all, are synthetic statements.Hence, the middle section of this chapter described the value of empirical researchabout various types of exceptionally talented individuals. It describedbroad categories of empirical research and discussed the origins and uses,in particular, of representative techniques of quantitative research.The middle pan of the chapter concluded with a discussion of theinterpretation of empirical resultsrecognizing the assumptions of empirical researchand the applications of empirical research to the educationof gifted students. The final section of this chapter discussed the treatmentof talent in the public schools. Discussion here contrastedthe notions of rarity and universality with respect to the identification oftalent, the practical methods for dealing with talent in the schools,and the related issues of testing error and equity. Inparticular, two empirically based ways to achieve equi-

34 28 Tim Study of Gifted Shields ty in programs for gifted students were mentioned. The discussion also considered the twin issues of elitism and anti-intellectualism in gifted programs, and suggested ways in which that dilemma might be resolved. Fi- nally, the need to develop talent was related not only to improving gifted programs, but to improving schools in general. Heredity and Environment

I.Focusing Questions II.The Nature and Nurture of the Debate about Nature and Nurture A.Research Difficulties 1.Difficulties in accounting for sources of variability in IQ 2.Practical value of the research question III.Research on the Nature versus Nurture Question A.Twin Studies 1.Alternate interpretation of findings from twin studies 2.Cyril Burt's influence 3.Problems with the research design B.Parent-Child Studies 1.Methodological problems 2.Reanalysis of early studies C.Group Differences and Studies of Test Score Differences 1.Jensen's research 2.Individual and group differences 3.Coleman's research IV.Historical Identification Trends A.Brain Research B.Psychometric Testing 1.Local norms and quotas V.Misconceptions in Hereditarian Interpretations of Research A.Heritability versus Inheritability VI.Alternate Conclusions VII.Summary

36 27 VI Herta, aid Entire: -.weld

Focusing Queslions 1.Is there a difference between artificial early giftedness and real giftedness? Describe the sense in which the adage -early bloom, early rot* is true. 2.What sorts of studies have been done to investigate the degree to which levels of intellipnce can be attributed to genetic or environmental causes? Mmt are some of the specific problems of these studies? Some of the general problems? 3. How is it possiblein both hereditarian and environmentalist perspectivesfor children to be more intelligent than their parents? What evidence supports this phenomerm in each case? 4.If the source of all variability in individuals' !Qs could be attributed to genetic cause, could we therefore infer that group differences in IQ were caused by genetic differences? 5. How are the preceding 'questions related to the identification of giftedness in minority children? What advantages and disadvantages are associated with alternative methods of identifying minority children? 6.If giftedness is not genetically determined, are gifted programs defensible? Why not, or in Miat way?

THE NATURE AND NURTURE OF THE DEBATE ABOUT NATURE AND NURTURE

The termsof the nature-nurture controversy form the subject of this chapter. Discussion considers the hereditarian and environmental views about the relative influence of genetic endowment and social circumstance on the acquisitionof intelligence.This controversy is important to gifted educators for several reasons:

Identification of giftedness most often depends on high IQ scores. Few black, Hispanic. and American Indian children have been placed in gifted programs. Elitist conceptions of giftedness rest heavily on the belief that intellectual ability is both genetically determined and unchangeable. One's view of the controversy leads to certain conclusions about the degree to which talent development 6 possible.

Whether gifted children are born gifted or whether they are gifted because of their environment has long been a matter of controversy. Most teachers and parents in the United States, however, assume that children's intelligence is inherited. In fact, they tend to see obvious environmental advantage as inducing an artificialkindof ability, not real giftedness. They may, for example, believe that kindergarten students whose parents have helped them learn to read display an artificial ability. They

:17 Heraity and Environaievit 29 often predict that such artificial abilities will disappear as thespecial conditions of the home are replaced by the normalconditions of school. They see the environmentaladvantages as a kind of crutch that truly bright children can do without. This viewpoint sometimes appears to be confirmed byexperience. To a certain extent, theadage "early bloom, early rot" is correct. School can equalize children's abilities, and in the case of giftedchildren, the equaliza- tion means a drop in achievement. The levelinginfluence of a curriculum designed for the average student often does have anadverse effect on gifted students' achievement, regardless of earlyadvantages in their environment. Research Difficulties This issue is troublesome. There appears to be no feasibleexperimen- tal design that can answer with certainty thequestion of the relative influence of heredity and environment onhuman cognitive behavior. In laboratory studies of animals, geneticallyrelated individuals can be raised in carefully controlled environments, but the environmrcannot be controlled very effectively in experiments with humans. 7 .,,studies to determine the major influences on human intelligence cannot sort outthe effects of environment from those of heredity.Scientists, of course, keep trying even in the face of this difficulty. Theircontinued efforts attest to the significance of the question. Difficulties in accounting far sources of variability in N.The social implications of the issue of nature and nurture aresuggested by the position of scientists on the controversy, whichvirtually always reflects their politics and social status. One study (Pastore, 1949)found that among 24 scientists, 11 of the 12 liberals were environmentalistsand 11 of the 12 conservatives were hereditarians. Another study(Sherwood & Nataupsky, 1968) showed a relationship between opinion onthis issue and socioeconomic factors. Of over 80 scientists studied, thosewho had been the first-born, who had grandparents born in theUnited States, who had parents with higher educationallevels, who had higher grades in college, and who had grown up in rural communitiestended to interpret intelligence test scores in a hereditarianfashion. As Gage and Berliner (1975) point out,the results of these studies weaken researchers' claims of scientificobjectivity on this controversial issue. Gould (1981) carries this point evenfurther in his book, The Ms- :measure of Man. Heprovides examples not only of biased results, butof fraudulent results used to support hereditarian arguments.(If the advocates of environmental sourcesof superior intelligence have fudged their results as much as the other side, they havebeen very fortunate in not getting caught.) SO Herodity and Environment

Practical value of the research question.Researchers have repeatedly tried to discover how much variation in individual behavior on IQ and other academic tests is explained by hereditary factors and how much is explained by environmental factors. Although this question is not one that can be answered without qualification, the literature includes awide range of estimates. Arthur Jensen, perhaps the most well-known proponent of the hereditarian view, opted for an 80 percent heredity figure in his 1969 article, "How Much Can We Boost IQ and Scholastic Achievement?" Christopher Jencks (Jencks, et al 1972), an equally well-known proponent of the environmental explanation of intelligence, estimated the heredity figure to be about 45 percent. Some researchers believe that the percent contributed by heredity may be even smaller (e.g., Lewontin, Rose, & Kamin, 1984). Some, on the other hand, (e.g., Hunt, 1961) don't even think we should ask the question. They have recommended that, instead of invest- ing resources in attempts to make better guesses, specific influences should be investigated. Such investigations have been undertaken, but they are not likely to get the same attention that studies of the more controversial question get. The broader studies deal with an ideological question that reflects predominant views of the mainstream culture.

RESEARCH ON THE NATURE VERSUS NURTURE QUESTION The types of research studies that have been used to argue for innate differences in the intelligence both of individuals and o different racial or ethnic groups are (I) comparisons of twins' and other siblings' intelligence test scores, (2) comparisons of intelligence test scoresof adoptive and natural parents with the test scores of their adopted and natural children, and (3) comparisons of the test scores (both IQ and achievement) of different racial and ethnic groups with the scores of the white majority.

Twin Studies Studies of identical and fraternal twins have served as a mainstay fou research on hereditary versus environmental factors in cognitive development for over a century. Gahon's Inquiries into Human Faculty and ItsDrvel- oinnent (1883) included studies of monozyptic (identical) twins anddyzygo- tic (fraternal) twins. As one of the most eminent English scientists of that time, and a staunch proponent of heredity as the exclusive source of human abilities, Calton influenced the thinkingof both English and Ameri-

3 9 Hetodity and Environment 31 can scientists. One of the most prominent amongthem was Cyril Burt, whose twin studies formed the foundation for even later studies, such as those of Arthur Jensen. Bun's results,however, which supported a hereditarian view of intelligence, have been discredited because they were falsified. Most twin studies have found that the strength of the correlation between IQ scores tends to increase with both genetic and environmental similarity. For identical twins raised together, the correlation coefficient is, on the average about 0.85. When identical twins are raisedseparately their IQ scores tend to correlate about 0.75 (Gage & Berliner, 1975). The fact that both of these correlations are higher than correlations between either fraternal twins or nontwin siblings has been used to argue that heredity is more important than environment. The argument proceeds as follows: If identical twinswhose heredity is identicalare more alike in IQ test scores even when they are raisedin different environments than fraternal twins raised in the same environment, then heredity must influence IQ scores more thandoes environment. Kamin (1981), however, suggests alternateexplanations for the twin study findings.

Alternate interpretation of findings from twin studies.According to Kamin one reason identical twins raised together show a higher correlation of IQ score than fraternal twins is that identical twins are treated similarly and tend to spend a great deal of time together. In other words, their environment is more similar than that of fraternal twins and nontwin siblings. Kamin found that the IQ scores of identical twins whose parents reported that they had tried to treat them the same were higher than between identical twins whose parents reported that they had not tried to treat them the same. In further support of his argument, Kamin notes that the similarity of IQ scores varies with gender, whereas the similarity in height does not. Female twins' IQ scores are more similar than male twins' scores. Kamin believes that this result is obtained because female twins are usually closer together more of the time (for example, they usually sleep in the same room) than are male twins. There is no gender difference in correlations of the height of female twin pairs and male twin pairs, however. Kamin thinks this different correlation pattern stems from the fact that height is highly heritable and IQ score is not. On the other hand, nomwin siblings are treated less similarly than identical or fraternal twins. Moreover, their environments are less similar: they are born at different times and have different age-mates, for example. Kamin claims that these differences, not different heredity, explain the lower correlation of non-twin siblings' scores.

4 0 ft ihreiity and iswirossamed

Cyril Burt's isijistersee.Burt's discredited study (e.g., Burt, 1966) is the largest study correlating the IQ scores of separated identical twins. This study formed the foundation of many hereditarians' belief in the profound influence of heredity in determining IQ score. Cyril Bun's articles were supposedly based on 53 twin pairs. Ti. IQ correlation of these separated twins was very high, higher than that reported in other studies. More significant, however, was his assertion that the environments of the separated pairs were totally dissimilar. This meant that their similar scores were due only to similar heredity. According to the authors of Not in Our Genes (Lewontin et aL, 1984), there are several objections to condusions favoring heredity as a factor in the high correlation of IQ scores of separated twins. Lewontin, Rose, and Kamin cite the ready acceptance of Burt's study by other professionals as a reason to doubt the hereditarian thesis. According tothese authors, such ready acceptance raises questions about the objectivity of the scientists who incorporated Burt's results into their work. They contend that the implausibility of Burt's reports should have been noted at once: Burt failed to provide a description of how hecollected his data, when it was collected, or where it was collected. Hedid not even name the IQ test he used. More criticalreaders observed that some of Burt's correlations were "impossibly consistent." Further investigation led to the discovery that Burt had collected no data at all during the time periodwhen he was purportedly studying separated twins. Once Burt's results were impugned, so were the results of those who had based much oftheir argument on his findings.

Problems with ihe research design.Lewontin, Rose, and Kamin (1984) also raise questions about the validity of all separatedtwin studies. Weakening the validity of other such studies is the fact that the research design yields inevitably biased results. Twins who have trulybeen raised apartthat is, separated from birthmay not know of eachother's existence. These twins cannotrespond to the appeals of scientists for volunteers. Inspection of the data on most"separated" twins reveals that they were not actually raised separately. For example, Shields's (1978) study included "separated"twin pairs that had been reared in related branches of the samefamily. There were only 13 pairs in which the two twins had been rearedin unrelated families. The mast common pattern shown in the study wasfor the biological mother to rear one of thetwins and the maternal grandmother or aunt to rear the other. The IQ correlation of the pairs reared inthe same family network was significantly higher than the correlationof the pairs reared in unrelated families. This difference is an environmentaleffect because the twin pairs are genetically identical. Also, there isprobably an environmental influence even on the twins who werereared in unrelated families

4 1 Heredity asd Eavirmsorst 33 because they were reared by closefamily friends. The Shields study required only that at some time duringchildhood the twins had been reared in different homes for at least five years.Many of their subjects had spent most of their lives together.

Parent-Child Studies The other mainstay of research on geneticand environmental sources of intelligence is the study of correlationsbetween the IQ scores of adopted children and the scores of their natural andadoptive parents. Most of the research from this approach shows highercorrelations between children and their natural parents than betweenthe children and their adoptive parents. However, Kamin (1981)questions the results of these studies. He notes problems with thesocioeconomic status (SES) classificationsand reports that the Munsinger(1975) study, one of the most famous of recent studies, has been discredited and shouldbe discarded.

Methodological problems. There are someproblems inherent in the design of all of the studies of adoptive parentsand their children. First, adoptive parents constitute a special groupthat, in general, differs on variables that are not adequately "captured bydemographic measures of environment" (Kamin, 1981, p. 118). Thisrestricted range of environmental variance alone can account forlower correlations between adoptedchil- dren and their adoptive parents thanbetween adopted children and their natural parents. Kamin cites improvedstudies (e.g., Horn, Loehlin, & Willerman, 1973; Scan & Weinberg,1976) that suggest a higher correlation betwecn the IQ scores ofchildren and their adoptive parents. f he improved designs and improved statisticalanalyses of these more recent studies result in a significantly lowerestimate of heritability than earlier studies; so low, in fact, that, according toKamin (1981), even behavior geneticists who conduct adoptionstudies have begun to point out th A -.ome of the data suggest zero heritability.

Reanalysis of early studies.McAskie and Clarke (1976) have reanalyzed data from two major earlystudies, Snugg (1938) and Skodak and Skeels (1949). In these studies,adopted children were found to havemuch higher scorf!s than their natural parents.The reanalysis of these data also produced quite low correlations between parentand child IQs. Snugg reported a regressionof 17 points toward the meanfor a sample of 312 foster children. Averagenatural parent IQ computed from the Snugg data was 78, whilethe mean IQ of the offspring was95. The correlation of parent-offspring IQ was0.13. In the Skodak and Skeels data, averagenatural parent IQ was 86, and mean offspring IQ was106, a regression of 20 points,whkh overshot the 42 34 Heredity modEnvironment mean. The correlation of natural parent-offspringIQ was 0.38. In addition, three of five comparative cohorts in the original study regressed from 1 standard deviatiun below the mean (parent IQ) to I standarddeviation above the mean (offspring IQ). Regression is not predicted by the genetic model in the case of average parent IQ (McAskie & Clarke, 1976), and these extreme differences in the scores of children and their natural parents argues for a large environmental influence onintelligence. McAskie and Clarke (1976) also reported the very low correlation coefficient (r0.08) of Terman and Oden's comparison of the IQs of their gifted subjects with the IQs of their parents. The mean parent IQ was 152 and the mean child IQ (parent and offspring were of the samechronologi- cal age at testing) was 133. A restricted range helps to explain thislow correlation, however. When we consider that Terman and Oden (1959) argued that their subjects provided a relatively homogeneous environment for theirchildren, this figure is still surprising. Genetic theory says that as theenviron- ment becomes more stable, or homogeneous,the proportion of variance attributable to genetic causes becomes greater. A high correlation between IQ scores of parents and their children cannot answer thequestion of whether the correlation reflects environmental or hereditary factors.Given valid and reliable tests, however, a low correlation is evidenceagainst strong hereditary influence.

Group Differences and Studies of Test Score Differences Do different groups of students display systematic differencesin IQ? Can such differences, if they exist, be ascribed to differencesin the genetic background of the groups? These questions are among those mostoften asked in the nature versus nurture debate. These questions arestill asked, despite the fact that the popular hereditarian view is nowknown to rest on shaky findings (i.e., Burt's discredited studies). In addition, asGould (1981) points out, even if IQ differences in individualscould be ascribed to genetic causes, it would still be difficult to ascribe groupdifferences to genetic causes. Samuda (1975) statesfour premises alxmt IQ tesrs that limit the validity of any hypothesis about racial differences in intelligence:

The scientific basis for the existence of racial differences inintelligence rests on four premises. (I) The conceptof intelligence is well understood. (2) Intelligence tests provide a true and reliable measure ofintellectual furwtion- ing, and the difTerences in scores reflect differences inintellectual poten- honwgeneous tial....(3) Both whites and blacks represent separate but gri m ps racially distinct. (4) The effectsof the variance in physiological, so- ciocultural, and economic conditions of both groups can heignored. (p. 40)

4 3 Heredity mid Envirmsent 35

These unsound premises are not just theprovince of hereditarians. Some researchers who subscribe to environment as amajor influence in achievement subscribe to all but the last of thesefour unsound premises. Usually such researchers attribute the meandifferences in the IQ scores of black and white children solely to thedebilitating effects of minority attitudes and habits. Hence, the investigationsof intellectual inequality (e.g., Coleman, 1966; Mosteller & Moynihan, 1972; Passow,Goldberg, & Tan- nenbaum, 1967; Williams, 1970) attempt todiscover those features of the culture of the disadvantaged that makethem appear intellectually deficient. Such analyses are more cogent,however, if they acknowledge that disadvantage may be a function of the wayin which "disadvantaged" groups are treated by"advantaged" groups.

Jensen's research.The influence of disadvantaged statushas been reported even by hereditarians. Jensen (1969),fbr example, found that socioeconomic status (SES), not racial or ethnicorigin, produced IQ differences. Of course, Jensenbelieves that differences in intelligence are inherited. Hence, he believes that the poverty ofblacks is a result of their lower intelligence. Those who believe the environmenthas the greater influence usually advocate the opposite causal relationship,that is, that the low intelligence test scores are the effect of poverty. In his research, Jensen also found thatchildren from middle-dass families earned higher scotes on tests requiringreasoning than did poor children; he observed less difference in theirability to memorize. Jensen cites the difference in reasoning scores to supportthe hereditarian position. The difference in reasoning ability,however, can be explained as a result of differences in schooling. Lower-SESchildren are schooled differently from middle- and upper-SES children(Anyon, 1980; Oakes, 1985; Wilcox & Moriarity, 1977); and one of thesedifferences is that wealthier children are encouraged to reason and toconsider the implications of their actions, whereas lower-SES children areencouraged to do what they are told without questioning authority (e.g.,Wilcox, 1982). Jensen (1973, 1980) reports thatwhite-black differences are slightly larger on nonverbal tests than on verbal tests.Jensen (1980) also reports that the magnitude of the white-blackdifferences depends on the extent to which these tests reflect general intelligence(i.e., in the vocabulary of factor analysis, the extent towhich the verbal and nonverbal tests "load" on the general intellectual factor, g). Jensenincorrectly assumes that the om- nipresence of g in factor analyses of testsindicates that it is a real psychological factor. There is more reason to interpret g 3S acharacteristic of the tests and the statisticalmethods of analysis than of thepsychological makeup of those who take tests (Gould,1981).

4 4 36 Heredity r4d Eivirouskest individual and group differences. The controversy about groupdifferences in intelligence is long-standing. However, it derives from amisun- derstanding of the implications of research findings on individualdifferences in intelligence and a misunderstandingof the nature of IQ tests. Lewontin, Rose, and Kamin (1984) try to clear up the confusionby likening IQ test scores to the growth of corn seed:

Suppose one takes from a sack of open-pollinated corn twohandfuls of seed. There will be a good deal of genetic variation between seedsin each handful, but the seeds in one's left hand are on the average nodifferent from those in one's right. One handful of seeds is planted in washed sandwith an artificial plant growth solution added to it. The other handful isplanted in a similar bed, but with half the necessary nitrogen left out. When theseeds have germinated and grown, the seedlings in each plot are measured,and it is found that there is some variation in height of seedlingfrom plant to plant within each plot. This variation within plots isentirely genetic because the environment was carefully controlled to be identical for all the seedswithin each plot. The variation in height is then 100 percentheritable. But if we compare the two plots, we will find thatall the seedlings in the second are much smaller than those in the first. This difference is not atall genetic, but is a consequence of the differencein nitrogen level. So the heritahility of a trait within populations can be 100 percent, but the cause ofthe difference between populations can be entirelyenvironmental. (p. 118)

Hereditarians often fail to understand that the causes of thediffer- ences between racial, ethnic, orsocioeconomic groups are not the same as the sources of variation within the groups. Ifthe ensironments of the groups are truly differe at, thenthere is no way to reason from a condusion of genetically determiaed individual differences to aconclusion of genetically determined group differences (Lewontin et al., 1984).

Coleman's research.Anothe1 way of looking at the differences between the school performance ofchildren irom different groups is found in Coleman's (1966) Equality of Educational Opportunity.This study reported on students' achievement acrossthe United States, and it found that blacks, Mexican-Americans, and lower-SES white students scoredconsiderably lower than middle- and upper-SES whitestudents in almost all school situations. What has commonly been identified as mostsignificant about this finding, however, is that the quality of theschoolsquantifire by measures such as teachers'salaries, size of library, level of fundingdid not seem to affect studentachievement significantly. The report, however, suggested something quite different: that the qualityof the schools (measured in dollars expended) was, in fact,nearly equal, even though blacks and whites were consistently segregated. Ryan (1976) takes exception to the controlvariables in the Coleman study, however. He says that controlling forfamily background or so-

4 5 Herta, and &winnow/vet 37 cioeconomic status (SES) before comparingdifferences in children's achievement as a function of differences intheir schools eliminatedwhat should have been a major variable studied.It is not surprising thatCole- man's findings were that school resourcesmade little difference inachievement; he did not comparewhite upper-class schools with whitelower-class schools. His comparisons removed thesignificant variable of SES. Ineffect, he compared the achievement ofupper-SES students fromdifferent schools, and the achievement oflower-SES students from differentschools. It is no wonder then, according toRyan, that differences in achievement bore little relationship todifferences in the schools. There was very little difference between the differentschools attended by white, upper- class children. In controlling forchildren's family background, Coleman was, in effect,eliminating the variable of greatest interest.His study simply found that there is little difference in theeffectiveness of schools attended by children of the same socioeconomicclass. He certainly did not find that thereis little difference in the effectiveness of schools attendedby working-class children versusthose attended by upper-class children. This latter conclusionis, nonetheless, the one most people draw from his research. Recentanalyses also confirm the observation that, once SES is controlled,expenditures are an ineffective measure of the quality of schools (Walberg& fowler, 1987).

HISTORICAL IDENTIFICATIONTRENDS

Discussion of group differences inIQ focuses on the comparisonof blacks and whites. In the context of the manypossible comparisons that could be made between ethnic and racial groups,this focus is revealing. The characteristics of Jews, Asians, Afro-Asians,Irish, Slays, and a variety of other "racial" groups could and, at onetime or another, have been scrutinized with equal zeal. The Immigration Restriction Actof 1924 was based on studiesby Goddard. Yerkes, and others. Butin the United States no comparisonhas received the persistent attentiongiven to IQ differences betweenblacks and whites. The most consistentfinding in the literature on black-white IQ differences is that the mean scorefor the black population is about one standard deviation below the white mean.For the population measured generally, this difference seems to havebeen constant for some time. That such differences have vanishedfor some ethnic immigrant groupsindi- cates the intransigenceof the problem that assimilationinto the mainstream has posed forblacks, at least in part because of the extmntdiscrimi- nation they have suffered at thehands of whites in the United States. Indeed, it is only recently that theblack literacy rate has approachedthe white literacy rate in the UnitedStates.

46 38 Heredity and Enviroseeent Brain Research

The earliest attempts to test intelligence reflected thehereditarian views of the researchers. They predicted that there would bea high, positive correlation between intelligence and brain size. Paul Broca was a nineteenth century French physiologistwho studied the brain. Broca identified thearea in the brainnow referred to as Broca's areathat controls speech. Broca's method for testing intelligence, however, was less reputable than his physiological studiesand, consequently, is less well known. His approach was to use the skulls of deceasedmen and women, some noted scholars, some laymen, andsome criminals, and compare the sizes of their brains. Broca expected to find that larger brainswere to be found in the skulls of male scholars (see Box 2-1). His theorywas not borne out by the data, however. As Gould (1981) pointsout, some of the largest skulls Broca measured belonged to undistinguishedpersons, even criminals. This method of research did not readily confirm the hypothesesof the investigators, and less crude methodswere developed. Galton, for example, tried testing sensory and motorresponses such as reaction time. This and similar methods also failedto confirm the hypotheses of the researchers.Ironically,it was the psychometric method ofa non- hereditarian, Alfred Binet, that provided themeasurement tool that lent itself most readily to the hereditarian hypothesis.

BOX 2-1Women's Brains

Broca's predilection for measuring the size of skullsarose from his desire to test a popular hypothesis: that white, northern European malesare more intelligent than other groups, particularly blacks, and particularly women, white or black. Broca believed that women were not as smart as men becausewomen had smaller brains. He knew, of course, that part of the reason women had smaller brains was that, in general, they had smaller bodies than men. He claimed, however, thateven taking body size into consideration, men were smarter and their brains larger than body proportion couldac- count for. His reasoning, however, was circular. Gould (1980) quotes Broca:

We might ask if the small size of the female brain depends exclusivelyupon the small size of her body. Tiedemann has proposed this explanation. Butwe must not forget that women are, on the average, a little less intelligent than men, a difference which we should not exaggerate but which is, nonetheless, real. We are therefore permitted to suppose that the relatively small size of the female brain depends in part upon her physical inferiority andin part upon her intellectual infenority.

Many other scientists of the time agreed. One, in particular, was *horrified" (Gould, 1980, p. 155) at the idea of offering women an opportunity for higher education. He proclaimed the idea not only misguided, but dangerous to society. Though these ideas now seem ridiculous to most of us, we should not forget that they were accepted wisdom less than 100 years ago.

4 7 Heredity wed Enviresamit 39

Gould reanalyzed Broca's data comparing the skull capacity of menand women. Using a statistical correction for the effectsof height and agg, he found that there is still an average difference of 113 grams. However, ashe points out, his formula could not take into account cause of death, which often has a significant influence onbrain size. There is, for example, a 100-gram difference betweenthe brains of persons who have died of degenerative arteriosclerosis and those who died by violence oraccident. Many of &ma's female subjects were elderly; and according to Gould,it is likely that lengthy degenerative disease is an explanatory factor in their relatively smallbrain size. The best means of correcting for body size is still not certain. Onenineteenth-century scientist, who argued that muscle mass and force shouldbe considered, found that when he took this factor into account, women's brains wereproportionately larger than men's. The value in Gould's retelling of events in this period in sciencehistory is in recognizing the obvioususe of science to confirm commonprejudice. Proponents of white male superiority claimed that women and blacks woelike children, and, consequently could not be afforded rights equal to those ofwhite men. Gould takes issue, not just with Broca's conclusions and those of his modern counterparts,but with "the whole enterprise of setting a biological value upon groups. . (o. 159). He regards such science as both irrelevant and injurious. -...._. ,

Psychometric Testing Binet, who developed the first psychometric intelligence test,thought intelligence was too complex to be represented by asingle number (Gould, 1981). He cautioned users of his test that the test score was an averageof performances on the various tasks in the test, not representativeof an innate psychological entity. Binet had not designedhis scale to confirm a hypothesis about innate ability; he designed thescale to identify children who might need special instruction in order tosucceed in school. American psychologists who imported Binet's testand popularized it disregarded Binet's cautions about his test, however.Hereditarians like H. H. Goddard and Lewis Terman promotedthe test as an index of innate intelligence. Goddard used the IQ test to identify personswhose intellectual inheritance was "poor" in order to segregatethem from the general population and to keep them from havingchildren (Gould, 1981). Terman was primarily interestedin studying the development of children whose intellectual inheritance was, according to histheory, superior to that of other children. Terman revised Binet's scaleand developed the Stanford- Binet Intelligence Scale. He designed this test forresearch on the development of gifted children. In order to identify subjects for his longitudinalstudy of gifted children, Terman (1925) administered his IQ test tochildren who were referred for testing by their teachers.His approach, say critics, limits the generalizability of the findings to gifted childrenfrom middle- or upper- dass families. 48 40 Heredity aped Environevest Terman's method of selection (teacher referral) yielded a sample in which the representation of working-class and underclass children is low. Ten times as many subjects came from professional families as wouldbe expected from the percentage of professional families in thepopulation (Hildreth, 1966). This finding did not seem anomalous to Terman; rather, it confirmed his belief in the importance of inherited intelligence. From the 1920s to the late 1960s, teacher referral followedby IQ testing was the basic approach used to identify gifted students. In the late 1960s and early 1970s, the effort to identify economicallydisadvantaged gifted children from ethnic or racial minority groups was a federal priority. Nonverbal intelligence tests were recommended as alternatives tothe standard IQ tests, and the creativity tests developed during the late1950s and early 1960s were also used to identify gifted children. Checklists itemizing behaviors associated with giftedness were given to teachers asguides to their referral of disadvantaged children. These approaches are still common provisions foridentifying minority gifted children. These approaches have problems, however,associated with (1) their lack of predictive validity and (2) theirunreliability. There is little evidence that the disadvantaged children identifiedby these methods are those children who have the greatestacademic potential and who, on that account, most need special attention. Behavior checklists, especially, are of dubious worth. Most ofthem are so subjective thatthey make it likely that the children who are identified are those who are mostattractive to teachers, rather than those with unusual potential for superior academic or artistic achievement.The most promising methods for identifying gifted minority children,establishing local norms or using quotas, are seldom used because theyrequire greater effort to implement.

Local norms and quotas.Itis more difficult to establish local norms or to identify the top 3 percentof a specific population than it is to compose a behavior checklist oradminister different tests as alternatives to IQ tests. Failure to use reliable identification methods is also afunction of the dis- repute into which IQ tests, inparticular, and standardized tests, in general, have fallen. Only a few efforts have been made to use IQ tests tobenefit disadvantaged children. One of the most extensive efforts isMercer's Sys- tem of Multicultural Pluralistic Assessment(SOMPA, 1981), which is a form of reforming the tests by which gifted children areidentified. By using local norms, however, educators can rely ontechnically adequate measures and at the sametime compensate for limitations in the use of IQ tests, such as those pointed out by Samuda (1975). AsSarnuda (1975) argues: It would be remarkable if there were no suchdifferences in the scores of different groups on test results, for some aspects of the valuesand content of tests areto a lesser or greaterextentnecessarily alien to a large propor- 4 9 Handily cad Envisonsarat 41

tion of minorities. If tests are to be used forselection, if cutting scores continue to form the grounds for admission intoeducational institutions or to the avenues of access to employmentopportunities, then it follows that without differential points of entry rninorizies will remain at adisadvantage, equity of educational and employment opponunities will bedenied them, and the gap between the dominant majority and the strivingminorities will increase. (pp. 129-- l 30) Unfortunately, the question of identificationof gifted students is bound up with the question of IQ heritability.The practical need to find those students who can accelerate through thecurriculum is confounded by the need to assert these students' innatesuperiority. Establishing the proper perspective fromwhich to view identification efforts requires a sensible critique of the hereditarianviewpoint.

MISCONCEPTIONS IN HEREDITARIAN INTERPRETATIONS OF RESEARCH

Critics have pointed out the misconceptionsin hereditarians' understanding of the implications of even the most extremehereditarian case; that is, the assertion that intelligence is completelyhereditary. Central to their case is the concept of "heritability." Heritability versus Inheritability Heritability is a numerical estimatefrom measurementsmade in a single generationof the likelihood that agiven, genetically encoded trait, will be modified by the environment. Itis, in a sense, not a measure of inheritability at all. It is actually a measure of whatis left after environmental influence is accounted for. The moredifferent the environments of the subjects--or the more unstable their commonenvironmentthe lower the heritability estimate. Heritability is an empirical inference, not anempirical fact. In fact, estimates of IQ heritability vary widely.Researchers have, for example, computed rather low heritability estimates amongblack and low-SES populations (Scam 1981). Another problem in equating heritabilitywith inheritability is the questionable assignment of variables to"genetic" categories. One study (Scarr & Weinberg, 1976), forexample, classifies "mother's IQ" as a genetic factor; confounding the issue further,the mothers' IQs in this study were not obtained directlybut were extrapolated from the level ofthe mother's educational attainment. Despite this weakness,however, this study, provides evidence 9f the substantial effectof environment on the IQs of black children adopted by white families. Another important misconceptionabout the inplications of differences in test scoresis the notion that if IQ is inheritable, itis unchangeable. 5 0 42 Hmdity and Evãsw a trait is inheritable, however, it does notnecessarily mean that the trait cannot be affected by the environment. Moderngenetics has found that genic operation itself is responsive to environmental variation(Deutsch, 1968, p. 61). The history of intelligence testing and the types of decisions that have been based on IQ tests have reflected the assumption thatintelligence is both innate and fixed. The confusion of "inheritable" with"unchangeable" is characteristic of the hereditarian view and has had a continuingnegative effect on the opportunities of minority groups in the UnitedStates. This situation is perhaps more a result of how inequality isstructured in our nation than it is a result of research about group differencesin IQ.

ALTERNATE CONCLUSIONS Thus far, this chapter has reviewed the basic positions on therelative contributions of heredity and environment to intelligence. Ithas pointed out limitations in the popularhereditarian view. Chapter 11 discusses the impact of that view on minorities. We conclude that theimplications of the hereditarian view of intelligence are no longer tenable, asscienfists like Lewontin, Rose, and Kamin (1984) have pointed out. In our view, mankind is characterized by ageneral human intelligence that transcends boundaries ofracial and social order; this intelligence is made up of manyqualities that are reflected in such human abilities as dexterity, mimesis, language, abstraction,and metaphor. None of these intellectual endeavors has been describedwith sufficient precision by pedagogy, psychology, or biology to relate them to themechanisms of hereditary transmission. This view does not rule out the possibility ofdiffering genetic endowment. Differential developmentalstrengths have been consistently noted by such investigators as Bayley (1970) andFreedman and Freedman (1969). Thus it is permissible to speculate that anyhereditary differences between subpopulations, differences which mayeventually be confirmed by biology, will be quite specificallydelineated, whereas the general genetic rule will be that of structural andfunctional intellectual equivalence. Our position is speculative, however, It represents anhypothesis no more testa- ble than the opposing view: a view that has notbeen confirmed by study, despite the claims of hereditarians (cf. Gould.1981, pp. 273-274). The major implication of our view is that ifgifted children are gifted in large part because of their environmentalcircumstances, then (1) they can be made more giftedand (2) a greater number of gifted children could be created by a change of circumstances. Thekey concept in this interpretation is the notion of trait plasticity(Dobzhansky & Montagu, 1975) or cognitive modifiability (cf. Feuerstein,Miller, Hoffman, Rand, Mintzker, &

5 1 Herviity and Euvirowat 43 Jensen, 1981). Even conservative estimates suggest significant possible change. Scarr and Weinberg (1976), for example, suggest an average IQ "malleability" of 15-20 points. This degree of change seems to have been brought about without much conscious intent on the part of the families and schools of the subjects studied by Scarr and Weinberg. It seems reasonable to predict that devoted and systematic efforts to develop children's intellectual ability could raise this average considerably.

SUMMARY This chapter has identified the major issues in the nature-nurture debate on the acquisition of intelligence.Although the terms of this debate have not been established by gifted educators,its outcomes are critical to an understanding of gifted edutation. The chapter reviewed the empirical basis of the hereditarian position and critiques of that position. Results from twin studies, parent-childstud- ies, and studies of group differences in IQ were evaluated. The chapter also summarized the historical uses of intelligence testing and explained how the tests have contributed to the continuing debate aboutinnate human abilities. Finally, the chapter offered an alternate hypothesis about IQ heritability. This hypothesis is based on the notion of trait plasticity. Such an hypothesis takes accountof the social realities that confront many minority students, and it accords minority students greater access to gifted programs.

52 Intelligence Constructs

I.Focusing Questions 11. Normal and Abnormal Intelligence A.Ranking B. The Definition of Intelligence C.Theories of Intelligence 1. Behavioral theories 2.Single-factor theories 3. Multiple-factor theories 4.Dichotomous-ability theories 5.Information-processing theories D.Rarity: Normality and Abnormality 1.The value of rarity III.Adaptive Behavior A.Social Validation 1. Adaptation to peers 2.Adult acceptance of gifted children B. Cultural Pluralism and Ability C.Problem Solving as Adaptive Ability IV.School Performance A.Indices of School Performance 1. Achievement tests 2.Grades B. Learning Rate 1. Research on early mastery of academic skills 2.Research on speed of performance on academic tasks

44 latelkgrace Contracts 45

C.Family Support 1. Intellectual environment 2.Socioeconomic status and family support D.Attainment 1.Adult IQ of high-IQ children 2.Scholarship V.Intellect A.Western Cultural Tradition 1. The rise of literacy and reason B.The Role and Nature of Scholarship 1. Scholarship and scholastic aptitude 2.Progress and thought VI.School Responses to High-IQ Children A.Historical Overview 1. The influence of progressive education 2,The influence of Sputnik 3. Equal opportunity and back to basics B. Contemporary Practices 1. Identification 2. Program planning 3.Appropriate instruction VII.Summary

Focusing Questions 1.What is the relationship between a psychological construct such as intelligenceand psychological testing? Is intellectual giftedness merely that which is determined by intelligence tests? 2. In what critical ways do multiple-factor and dichotomous-ability theories of intelligence differ? Which do you think represent reality better? Why? 3.Explain the connections between practical problem-solving behavior and adaptive behavior. 4.Give four reasons why the achievement of gifted students is less extreme than their IQ. In your answer, be sure to show how these influences are connected to one another. 5.What evidence suggests that cultural and ethnic differences influence achievement? How are these influences related to socioeconomic status? 6.What differences did Terman and Oden (1947) find between the families of "underachievers" and "achievers"? What inferences do you draw from these findings? Defend your inferences, 7.In what ways is the construct *scholastic aptitude" a clearer guide to program planning than "intelligence"? 8.Describe seven justifiable program options for gifted students and show how they relate to the justification you have in mind.

5 4 46 intetkonice Constructs

NORMAL AND ABNORMAL INTELLIGENCE

This chapter examines views of intelligence. Psychological constructsof intelligence are of particular concern here because they have been used so widely to identify gifted students. Other constructsfor example, creativity, mathematical ability, leadershipare dealt with insubsequent chapters. We will examine the distinctionbetween normal and abnormal intelligence since to be "gifted" is sooften understood to mean "abnormally intelligent." The term "intellectually gifted" is also used as a synonymfor high-IQ students, hut the distinction between intelligence andintellect will also be discussed. The education of gifted students has historically depended on four premises: (1) intelligence exists, (2) substantial individualdifferences in intelligence exist, (3) the degree of these differences can be measured accurately among individuals, and (4) only some students possesssuffi- ciently exceptional intelligence to warrant altering the usualpractices of schooling. Over the years all of these premises have beenchallenged, and many of the challenges have had someeffect on opinion in the field, though their practical effect has been more limited. This sectionof the chapter considers the way in which giftedness reflectsranking by levels of' intelligence.

Ranking Our society prizes utility. We regard most highlythose persons, ideas, or inventions which are mostuseful. Scientists have developed empirical methods that can rank individuals according to the degree towhich they possess qualities that arebelieved to reflect social utility. The qualities that are regarded most highlyinvolve verbal and quantitative reasoning. Though these qualities are prized for their utility, they arealso considered to be intellectual qualities ofinherent worth. Our society also believes, like most societies, thatinherent worth is mirrored in social rank. This belief, however, actually has little todo with the inherent intellectual worth of an idea orintellectual work, and even less to do with the general humanworth (as opposed to the utility) of verbal or quantitative reasoning, wherever it may be found. Thisdistinction is an important one because it relates to what we expect ofindividuals said to be "gifted." We assume that individuals of high soda) rank make uselnlcontribu- tions to society. We also believe that as a result ofthese contributions such individuals should receive, or be entitled to receive, largefinancial rewards. The assumption works in reverse too: Failing tomake the distinction between social rank and the moregeneral human worth of thinking and reasoning, we erroneously conclude that those who reapthe greatest finan-

55 latenigence Contracts 47 cial benefits are by definition those who intrinsicallyhave the most to contribute. Not only do we conclude that wealthy businessmenand successful professionals are entitled to great wealth and largeincomes, we infer that they make an extraordinary contribution to society as aresult of their extraordinary capabilities. This belief seems quite reasonable, so it is difficult for somepeople to understand that apparent superiority might be the result of,rather than the cause of, high socialrank. The hereditary acquisition of wealth and privilege, however, not infrequently confers high social rank toindividuals who possess neither useful norinherently worthy qualities. Despite this important distinction, however, it is also truethat social rank itself gives many individuals the chance to improve"native ability." Hence, it is very difficult to separate the empiricalmethod of rank (based, for example, on verbal and quantitative reasoning)from the social context in which it takes place. Nevertheless, the empiricalapproach to the intelligence construct assumes thatthis separation is possible. Much of what occurs in schools reflects andstrengthens the process of social ranking. Although children may start out inschool with similar levels of ability. economically and socially privilegedstudents acquire increasingly greater benefits from school(Oakes, 1985). These bencfits are reflected in higher grades, achievement test scores, and even IQ scores(Rosenbaum, 1975). By the time these children leave school, they dohave skills that are more valuable to society. Hence,the practice of schooling inevitably perpetuates the scheme of socialrank. It is natural that this should be so, but it makes the ranking of inherent worth problematic. The authors of some intelligence tests, however,have purported to measure levels of inherentability with their tests. Some theorists arc convinced that these ability levels are not onlyinherent, but heritablethat is, that doing well on such tests can lx passed ongenetically. Others believe that, like grades, intelligence test scores reflectsocialized behaviors At the present time, no one really knowsthe degree to which these tests measure heritable traits orlearned behaviors. Our lack of certainty in these matters contributes to a persistent and oftenheated debate about the nature of intelligence and themeaning of its variability.

The Definition of Intelligence Frustrated educators say that intelligence is whatintelligence tests measure, and theircircular logic is intended as irony. Any definition of intelligence, however, is based on circular reasoningbecause intelligence is not a thing, but a construct. Theorists hold differing views of intelligence, in partbecause of their reasons for defining it.Piaget, for example, wanted to see human intelligence as an extension ofbiological adaptation. Hence, he defined intelli- 48 Inteffignsce Constructs gence in terms of the processes (e.g., assimilation andaccommodation) that help humans adapt to their environment. For Piaget, therefore, all humans act intelligently. Spearman, by contrast, wanted to define intelligence "objectively," which for him indicated a quantitative description. He viewed intelligence as a mental capacity that undergirds all cognitiveperformance, and his definition related amounts of intelligence to levels of cognitive performance. For Spearman, therefore, some humans act more intelligently thanothers. It is difficult to evaluate definitions of intelligence simplyat, correct or incorrect. Definitions can, however,be evaluated with respect to the way they are used. The discussion that follows considers five definitionsof intelligence and the social context in which they are embedded. These notions reflect (1) behavioral theories, (2) single-factor theories,(3) multiple-factor theories, (4) dichotomous-ability theories, and (5) theories of cognitive processing.

Theories of Intelligence

Most early theories of intelligence assumed that intelligencecould be measured. Son- t: early theories measured speed. Others measured the amount of a valued trait or constellation oftraits. Both approaches included tests that purported to measure the level of an individual'sintelli- gence. Individuals could, in the oneview, be more or less speedy. They could, in the other view, have more or less capacity. Many theorists,however, combined notions of speedand c-ipacity. The earliest quantitative definitions of intelligence related brain capacity to mental performance. The individuals with the largestbrains were assumed to be the most intelligent. Theorists of the mid-nineteenth century, however, were more interested in therelative intelligence of different groups of people than in differences betweenindividuals. Broca, Galton, and Morton measured the size of people's skulls to confirm the correlation between intelligence and brain capacity.For these researchers the results were disappointing, however, because white males did not turn out to have the largest skulls. Consequently,these men looked for different ways to measure intelligence. They wanted tofind methods of measurement that would confirm theirassumptions about the superiority of European males (Gould, 1981). Disappointed by the failure to correlate brain size with intelligence, Calton decided to explore the correlation between reaction time andintelligence. This approach was no morepromising than the measurement of brain -opacity. Galton did, however, contribute amemorable concept to the study of intelligence, the concept of statistical rarity. Psychometric tests such as the one developed by Binet and Simon proved more satisfactory than physical tests as a means torank the mental

5 7 bitenipace Co=from 49 performance of individuals and groups. These tests included sequencesof verbal, mathematical, and reasoning tasks. The tasks were similar tothe kinds of performance required in school. When Stern (1914) expressed as a ratio the relationship between taskperformance and age, he combined the idea of speed with that of capacity. This notionthe notion of "intelligence quotient"impliedthat the early mastery of skills would predict future capacity to master moreadvanced skills. Henceforth, learning rate was understood to be directly related to academic potential.

Behavioral theories.Behavioral psychologists emphasize observable phenomena and resist making inferences from behaviors. Therefore,they are more concerned with theobserved products of learning than with hypothetical mental abilities. This approach leads to the measurementof actual rather than potential attainment. Measures of adaptive behavior often conform to this theoretical approach. They are very helpful in determining the actual accomplishments of developmentally delayed children. They are much lessuseful, however, in describing the performance of gifted children. Highlyintelligent children may adapt or choose not to adapt for a variety of reasons.Gifted children may, for example, act immature in order to getattention or to blend in with other children. Whereas teachers want mentallyretarded children to adjust as well as possible to a "normal" sequence ofadaptive requirements, teachers should not want to make gifted childrennormal. Galton's (1869/1962) Hereditiny Genius was perhaps the first attempt to define intelligence in termsof its behavioral correlates. Gakon looked at what we today would call very complex behaviorsthe lifeaccomplish- ments of his eminent subjects. Recent methods of identifying gifted children rely onless complex indicators of behavior. Behavioral checklists (e.g., Renzulli,Smith, White, Callahan, & Hartman, 1977) attempt to relate observedcharacteristics to potential giftedness. They are troublesome because thesecharacteristics may not correlate verywell with aptitude for scholastic performance.

Single-factor theories.In the early twentieth century. Spearman (1927) postulated the existence of a global trait, g, that determinedintellec- tual performance. According to Spearman, differentindividuals had different amounts of g. Those individuals who had the most g wereable to perfbrm all mental tasks better than those who had less g, Spearman(1927) explained this idea in the following way:

[The] success of the same person throughout all variations of both form and subject matterthat is to say, throughout all conscious aspects of cognition whateverappears only tAplicable by some factor lying deeper than the phenomena of consciouciless. And thus there emerges the conceptof a hypo-

5 SO Intelligence Constructs

thetical general and purely quantitative factor underlying all cognitive performances.. g ...taken to consist in something of the nature of an "energy" or "power." (p. 5)

Based on this view, Spearman determined that any test of mental performance would reflect greater or lesser amounts of g. Wecsbler applied this view when he developed his intelligence scales. Wecshler (1958, p. 7) described intelligence as "the aggregate orglobal capacity of the individual to act purposefully, think rationally, and deal effectively withhis environment." There has been little empirical evidence to support Wechsler's claim, however. Researchers have found only a slight correlation between measures of intelligence and measures oflife success (Baird, 1985). They have, however, found a strong relationship between performance onintelligence tests and performance on academic tasks.Although this correlation does nothing to prove the merits of the single-factor argument, it does clearly establish the atiliiy of the tests that derive from this theory. The single-factor view of intelligence is also associated withheredi- tarian arguments, such as those put forth by Jensen (seeChapter 2). It is easy to draw an analogy betweensingle-factor IQ and a single, heritable biological trait, such as eye color or height. Intelligence tests seem toallow psychologists to measure the quantity of this trait in different individuals. This approach, however, is misleading. Intelligence tests measure complex behaviors that are defined in a social context. It isdoubtful that a single, heritable biological trait could account for the manyphenomena that are classified as intelligent behavior, even thatintelligent behavior described by IQ tests.

Multiple-factor theories.Because Spearman's single-factor definition did not account for much of the variability in anindividual's performance of different kindsof tasks, Thorndike and other theorists suggested a multiple-factordefinition. Thorndike (1927) proposed three kinds of intellectual capacities: abstract, mechanical, andsocial. Thurstone (1938) suggested seven primary mental abilities: verbal,number, spatial, memory, reasoning, word fluency, and perceptual speed. Guilford (1959) developed a three-dimensionalmodel of intelligence that accounted for 120 independent factors.The three dimensions of the model are operations, content, and products.According to Guilford, performance of an intellectual task depends onthe interaction of an operation, a content, and a product. Differentcognitive tasks require different kinds of interactions. A more recent multiple-factor view ofintelligence is described by Gardner (1983). Like Thorndike, Gardnerbelieves that intelligence gov- erns more than cognitivefunctioning. In fact, Gardner maintains that rl hdelligrice Comstructs 51 there are seven separate types of intelligence: linguistic,logical-mathematical, spatial, musical, kinesthetic, interpersonal, andintrapersonal. Among the multiple-factor theories, those that includefewer different abilities have the greatestpractical value. Like the tests derived from single-factor notions of intelligence, tests based on these nu1"rle-5'actor views relate to academic learning. At least several ofthe factors m .nany of these theories also relate to school performance. Forexample, Thurstone's tests of verbal and numerical reasoningpredict performance in specific school subjects because they sample behaviorssimilar to those expected in school.

Dichotomous-ability theories.Similar to the multiple-factor theories are the dichotomous-abilitytheories that hypothesize the existence of two distinct kinds of intelligence. These theories attempt toresolve the question of the influence of heredity as compared withthe influence of environment on intelligence. Some psychologists recognized that certainquestions on intelligence tests were dependent onchildrm's previous learning. Vocabulary items, general information items, and arithmetic items werethe ones most obviously sensitive to the effects of early learning.These psychologists believed that 1Q tests should include itemsthat are less sensitive to early learning. According to these theorists, suchitems would reveal native ability. Cauell (1971) distinguished between twokinds of intelligence: fluid and crystallized. According to Cattell:

Crystallized ability . operatelsi in areas where the judgments have been taught systematically or experienced before ...Fluid ability, by contrast, appears to operate whenever thesheer perception of complex relations is involved. It thus shows up in tests where borrowingfrom stored, crystallized, judgmental skills brings no advantage. (pp, 98-99)

Cattell's tests, the Culture-Fair Intelligence Tests(Cattell, 1950: Cat- tell & Cattell, 1960, 1963), made hisdistinction operational by linking verbal performance with crystallized intelligenceand by linking nonverbal (perceptual reasoning) performance withfluid ;liielligence. Cauell believed that the measurement of fluid abilitywould better represent the "true" intelligence levels of childrenfrom differentraces and socioeconomic backgrounds. Unfortunately, the resultsof his tests suggest that his assumption was incorrect.Measures of fluid ability have turned out to be no more "culture-fair"than measures of crystallized ability (Sattler, 1982). Kaufman and Kaufman (1983) attempted to make asimilar distinction with their notion of the differencesbetween intelligence and achievement. According toKamphaus and Reynolds (1984, p.215-216). the

6 52 Inteigesce Coutnitts Kaufmans viewed intelligence as the process of problem solving, whereas they viewed achievement as the "application of ...mental processing skills to the acquisition of knowledgefrom the environment." This distinction places "measures of what have traditionally been identified as verbal intelligence, . .general information, ...and acquired school skills" among the tests of achievement rather than amongthe tests of intelligence (Kamphaus & Reynolds, 1984, p. 215). The Kaufmans' notion of achievementclearly resembles Cattell's notion of crystallized intelligence. The Kaufmans based their test battery, the Kaufman Assessment Bat- tery for Children (K-ABC), on the distinctionbetween these two kinds of performance. The battery samples abilities that reflect both perceptual reasoning, or fluid intelligence, and acquired learning, or crystalliziintel- ligence. The test battery excludes behaviors that reflectcrystallized (or verbal) intelligence from its assessment of intellectual potential.The authors of the K-ABC consider this characteristic of their testsufficient to establish its culture-fairness. There is, however, littleevidence that minority students perform better on nonverbal than on verbaltasks Uensen, 1973; Samuda, 1975). Jensen (1973), for example, reports that mostblack children perform better on verbal tests than on nonverbal tests. Although the distinction between two kinds of intelligence has not been linked to differences in native and acquit td abilities,the verbal portions of tests based on this theory do predict scnool achievement.Whether or not we call such verbalperformance "intelligence" or "achievement" is moot; however, we do know that previous verbalachievement, like IQ, predicts future achievement.

Information-processing theories.In contrast to the psychometric approaches that emphasize the products of cognition, theinformation-processing therhists focus on cognitive processes. These theorists see intelligence as a sequence of events. The sequence includes phenomena such as arousal and maintenance ofattention, short- and long-term recall, and transfer of learning. Even if these processesdo condition intelligent behavior, they are so internal that theyresist measurement. Information-processing theorists study the influence of these mental events on the way that individuals solveproblems. They attempt to establish correlations between cognitive processes and moretraditional measures of aptitude, forexample, the ability to work formal analogies or mazes (Sternberg, 1977). Some of the distinctions made by information-processingtheorists resemble dichotomous-ability theories. According toSternberg (1982a), intelligence comprises the ability to acquire and to think in termsof novel concepts and conceptual systems.Intelligence involves the ability to assimilate systems that at first seem strange and then to applyexisting knowledge,

6 1 hiulligesce Colufrects structure, and skills to the new system. Hedistinguishes between analytic intelligence and synthetic creativity. In contrast to the dichotomous-ability theorists,however, Sternberg is not concerned with identifyingchildren's true levels of native ability. Rather, he is concerned with the ways that individuals learn.This approach is similar to that of Piaget (see Chapter 8). Because the theory of information-processinghas limited practical application to the measurement of intelligence,theorists have expanded the scope of their work to include metacognitive processessuch as those listed by Brown (1978). He identifies predicting,checking, monitoring, and reality-testing as executive processes of metacognition. Such processes are more easily measured thaninformation-processing, and they can also be taught. The relationship of metacognitive processes totraditional measures of intelligence and to academic performance isunclear, however (Stern- berg, 1977).

Rarity: Normality and Abnormality The notion of ranking assumes that there are differentlevels of p..r- formance. The idea of an age-ratio (i.e., Stern's concept)postulates differences in learning rate. Neither view, however, addresses the relationship betweenthe levels of performance and the numbers of performers.Gahon was the first to relate superior performance to its degree of rarity. Inhis view, the more a performance differs from the average or norm, the more rareit is. Superi- or performance is, therefore, akind of abnormality according to Galton. This statistical notion is crucial to the view thatintelligence is a characteristic that is normally distribukd in the population.IQ tests are constructed so as to confirm thisanalysis. Test authors include only those items that produce a normal distribution of scores in a representativesample.

The value of rarity. Whn intelligence tests areused to identify gifted individuals, they attempt to find those whoseperformance is most deviant. Typical school definitions of giftednessidentify the top 2 or 3 percent of students; Galton,though, was concerned with perfiirmance that was con.sitlerahly more rare. All efforts to identify superior performers arebased on the belief that particular kinds of high-level performancearesomehow--of particular value to society. They also assume that rareperformance is a "gift" that cannot be replicated.Therefore, individuals who possess gifts must be found, so that they can serve society. Nevertheless, individuals who demonstrate high levelsof intelligence may not be able orwilling to serve society. ln the highest range. intelligence scales cannot accurately predict sociallyvalidated performance. Within this

f; 54 isteiligrece Constnieu restricted range, they are not even good predictors of school success. (That is, among children with IQs above 145, it is not in general true that a child with an IQ of 180 will perform better in school than a child with an IQ of 150.) Although our society seems to value superiority, it values conformity at the same time. The exceptional academic performance of very high-IQ individuals may not correspond to the specialized needs of a pragmatic society. Moreover, the performance of individuals with extremely high IQs may be unpredictable. In the most extreme ranges, then. IQ scales maylose theirutility because deviant performance is unpredictable.

ADAPTIVE BEHAVIOR

The broadest definition conceives intelligence as adaptation to the environment. According to this view, the quality of intelligencediffers as environments differ. The demands of the physical or socialenvironment establish the terms by which intelligence is determined. School learning is only a small part of the modern social environment, so measures of aptitude for school learning consider only a narrow inimpretaiion of intelligence. This section of the chapter considers the broad concept of intelligence. It shows the relationship between adaptation to academic expectations and adaptation to other social demands. It also examines the social adjustment (i.e., adaptive behavior) of high-IQ students,

Social Validation A poll to determine how laypersons define intelligence (Sternberg, 1982b) found that it is commonly associated with success in anyendeavor except athletics, which was seen by mostof the people surveyed as not requiring extraordinary intelligence. This popular definition implies that intelligence is adaptive; and, in fact, the lay definition of intelligence is close to that of psychologists who have included in their definitions of intelligence a reference to its adaptive function (e.g., Piaget). Extraordinary intelligence, however, is sometimes seen as incompatible with adaptive behavior. For example, the view of intelligence as successful adaptation contrasts markedly with the stereotype of the maladap- five genius. Gifted children are often dtscribed as having no common sense and as being social misfits. In fact, the two views may represent judgments about adaptation to different environments: adaptation to school and adaptation to life. Be- cause adaptive behavior cannot bedefined without reference to the environment, what is adaptive behavior in one environment may be maladap-

6 3 beteiligewee CoNsfrects 55 tive in another. The advanced vocabulary associated with ahigh IQ score, for example, may represent a successful adaptation in afamily that values the articulation of ideas, but using such a vocabulary maybe maladaptive at recess. Just as intelligence can be defined narrowly asscholastic aptitude, adaptive behavior can also be defined narrowly. Whenpsychologists and educators use the term "adaptive behavior"particularly inreference to exceptional childrenthey usually meannoncognitive behaviorsthat help children get along in the classroom and in public.These behaviors include a wide range of skills, literally fromnose-wiping to making polite conversation. A particular social context is, in fact,essential to the definition of adaptive behavior. In general, adaptive behavior isbehavior that is validated by its congruence with social norms.Although most of us recognize that some chldren are "street smart," this type ofadaptive behavior is not acknowledged by social norms, and it will never betaught in schools nor measured by tests.

Adaptation topeers.Gifted children, at the elementary school level anyway, enjoy the acceptance oftheir classmates (Gallagher, 1985). Other children perceive gifted children as dressing "preppy,"wearing glasses, talking about boring things, carrying books, beingpolite, behaving well in public, treating others with kindness, doing well inclasses, getting good grades, and having common sense (Alexander,1985). At the preschool level, gifted children exhibit more social playforms than other children. They initiate play sessions with other children; andthey play more cooper- atively and share playthings to a greater extent thanother children (Bar- nette & Fiscella, 1985). At the high-school level, gifted students are atleast as well accepted, though not so popular as some other children(Coleman, 1961; Tidwell, 1980). Solano (1977) found ninth graders of averageability used equally positive adjectives to describe precocious students asthey used to describe themselves: determined, self-confident, sophisticated,serious, and cautious. Negative comments about gifted childrenincluded "dull," "conceited," and "self-centere J." Gifted children's acceptance by peers may be anacknowledgment of social distinctions. Gifted children's appearancedressing"preppy," (Al- exander, 1985) for examplereflects theirsocioeconomic status, a status that classmates acknowledge by their acceptance.

Adult acceptance of gifted children.Adults often attribute certain noncognitive traits to gifted children. These traitsinclude outgoingness, perseverance, a tendency todominate a group, and curiosity. These char-

f; 4 56 Into Ageism Constructs acteristics are probably more a function of the perceptions of children's middle-class backgrounds than of children's intelligence, however. Other studies of adult perceptions of gifted children tend to confirm this hypothesis. Teachers, in particular, are negative about any students who do not perform classroom tasks systematically, regardless of the children's intellectual abilities. Adults' positive perceptions about gifted students are always confounded by other variables. (For further discussion, see the section on school performance that concludes this chapter.) Similarly, adult perceptions of underachieving gifted students relate more to their views about underachievtment than to their perceptions of giftedness. They describe underachievers as lacking persistence in the accomplishment of ends, lacking the integration necessary to achieve goals, lacking self-confidence, and having feelings of inferiority (Terman & Oden, 1959). At least in part, these perceptions reflect impressions of the underachievers' socioeconomic status. The gifted children in Terman's study who grew up to be underachievers were, as a group, less affluent and their parents were less well echIcated than the average for the gifted group as a whole.

Cultural Pluralism and Ability Different cultural groups show different mean levels of performance on intelligence tests. When the effect of social class is not taken into account, blacks tend to score about one standard deviation lower on IQ tests than do whites. Whites, however, tend to score about two-thirds of a standard deviation below orientals. Less well known is research that identifies different aptitude patterns in different cultural and racial groups. Stodolsky and Lesser (1967) found that Jewish children score higher on verbal tests than they do on spatial, reasoning, and numerical tests. Blacks, too, score highest on verbal tests, according to Stodolsky and Lesser. Chinese and Puerto Rican children score highest on spatial tests, according to these researchers. Moreover, Stodolsky and Lesser observed identical patterns within the same ethnic group at both lower- and middle-class socioeconomic levels. Hence, these different aptitude patterns seem to be phenomena more closely associated with culture than with class. These different patterns of performance may reflect children's adaptation to the values of the cultures in which they are raised. Both the Chinese and the Jewish cultures, for example, have for many centuries placed a very high value on intellectual endeavors. The way in which such cultural legacies relate to findings like those cited previously is. however, obscure. The way in which such phenomena affect school performance is also obscure, though they may influence the development of cultural stereotypes, for better and worse. 65 Intelligence Constructs 57 Problem Solving as Adaptive Ability Successful people seem to be those who adapt best to their environment; adaptation, however, often involves solving problems.Though competence in most of the endeavors of life requires much more than academie problem solving, this section will consider only academic problem solving. Information-processing theory is relevant to the concerns of this section. Each of Sternberg's factors of intelligence is, in a way, construed as a category of adaptive behavior that seems to lie somewherebetween achieved skill and an ability to adapt new information usefully. Two factors that influence the success of problem solving are (1) the novelty of a task and (2) the degree to which a task is performed automatically (Sternberg, 1985). Ease of adaptation relates to the second of these factors. Gifted children, who are already familiar with many of the concepts and -kills required by grade-level work, perform school tasks more automatically than average ability peers. Their performance, therefore, is usually more accurate and rapid. Speed is important in many school tasks even though the correlation between perceptual speed and IQ scores is weak. The relationship between speed of reasoning processes and IQ, however, is moderate to high:0.70 for analogies. 0.50 for series completion, and 0.64 for classifications (Sternberg & Gardner, 1983). Other problem-solving strategies may also be adaptive in the classroom. Three separate processes that const.luteinsight may relate to the performance of academic tasks. These processes are:

1. selective encodingsifting relevant from irrelevant infmmation, 2.selective combiningcombining facts into a coherent scheme, and 3.selective comparisonmaking analogies between new information and previously acquired information. (Sternberg. 1985. p. 81) Within a school setting, it might appear that solving math problems correctly is the most adaptive behavior, and children with high IQ scores are adept at such adaptations.Readers should, however, remember that heading one's paper correctly, sitting still, and being polite are equally im- portimt adaptive behaviors in school; and gifted children varysubstantially in their willingness to meet such social obligations in the classroom. Sternberg (1985) uses the term "practical intelligence" to connote the kinds of adaptive social skills required for life success. These skills seem to entail the ability to interpret social situations effectively for one'sbenefit. Sternberg calls this ability "tacit knowledge." Again, this sort of adaptation was not the focus of thissection. Intelligence, as measured by IQ tests, is a mechanism for adapting to the cognitive requirements of school. It may not be amechanism for adapt-

66 58 beteftence Constructs ing to the noncognitive requirements of school or, forthat matter, of work (Baird, 1985). The next section of the chapter considers therelationship between high IQ and school performance.

SCHOOL PERFORMANCE

On almost any index of schoolperformance, students with high IQ scores. tend, as a group, to score above average. Their scores onacniemic achievement tests are as high as thoseof older students and adults. They are more likely than other students to earn high grades,complete college, ant" win academic honors. As adults, they tend to do wellin scholastic careers. A high IQ does not, however, ensuresuperior school performance, and a remarkable number (perhaps 15 percent)of gifted children perform at or below the averagelevel for children their age. Though theschotil performance of gifted children is affected bythe environment, it is, you may be surprised tolearn, also affected to some degree by the measures used to index school performance.The limitations of common indices of school performancegrades and groupachievement tests, in partkular often underestimate the achievementof high-IQ students. These low estimates provide falseevidence in support of a common misconception. This misconception holdsthat the educational needs of high-IQ children are not very different fromthose of other children and that spec;a1 provisions for them are unnecessary.Consequently, many high-IQ students are not provided theadvanced schoolwork they need in order to progress academically.

Indices of School Performance School performance is measured by indices (e.g.,grades) that, in many instances, cannot portraythe academic variable. iclevani to understanding gifted children. In fact, they oftenreflect variables that are irrelevant to academic progress.As a result, the contrast between a highIQ score and an average performance rating causesconfusion. The contrast creates doubt about the student's abilities and causesteachers to question whether or not the childwith a high IQ actually possesses superiorability. By contrast, the validity of the mostwidely used indices of school performance is seldom questioned. Grades aregenerally regarded by parents, teachers, andschool administrators asreliablfindicators of a student's mastery of academic material. It is rare that grades are recognizedfor what they arean unreliable, "gallimaufry" measure (Hunt, 1978, p. 109),confounded by differences in teachers' biases and skills and by myriadsociai conditionsof the student, lindligrowe Contralti: 59 the school, the family, and the community. Researchers, who do often doubt the reliability of grades, use achievement test scores; but achievement tests, especially group achievement tests, do not alwaysaccurately represent a gifted student's achievement either.

Achievement tests.The positive correlation between IQ and achievement test scores is usually construed to mean that innatelyintelligent children, who are identified by their high IQ scores, learn more readily than other children and so score higher on achievement tests. This interpretation is not necessarily accurate. IQ scores do not explain high (or low) achievement, despite their use as "diagnostic" instruments (Mercer, 1981). It is at least aslikely that students who score high on IQ tests also score high on achievement tests because (1) IQ and achievement tests measure similar cognitive skills and (2) the social variables that contribute to students' high IQ scores also contribute to high achievement test scores. Because some of the information required to score high on IQ tests is more generally available than that onachievement tests, IQ test scores depend less on students' in-school experiences than do achievement test scores. Children who have not yet receivedinstruction in particular academic skills, who have received poor instruction in them, or who because of emotional or physical handicaps have not benefited from instruction, are more likely to score high on IQ tests than onachievement tests. However, the differences between IQ test and achievement test performance are not all due to differences in the content of the tests. Difference in test format can result in discrepancies that have nothing to do with the child's knowledge of the content of the test.Although there are both group-administered andindividually-administered IQ and achievement tests, children identified as gifted areoften identified on the basis of individually-administered IQ tests. On the other hand, their achievement is typically measured with group tests. Perceptual, physical, or emotional problems keep a small proportion of these gifted students from demonstrating their mastery on group-administered achievement tests. However, because of the way these tests are constructed, even bright students who are not handicapped can fail to perform optimally on group- administered tests. Tests designated for a particular age level, for example, do not have enough difficult items to allow gifted students of that age to demonstrate their mastery of advanced skill, that is; such tests have a "ceiling" that is too low. Even among multilevel achievement tests, it is difficult to find tests with a high enough ceiling to assess accurately the achievement of gifted children twelve years old or older or of highly gifted children eight years old or older. In fact, the available individually-administered achievement tests also suffer from this latter limitation.Some researchers, such as 8 60 Intelligence Constructs Stanley (1976b), advocate the use of achievement testsdesigned for older students as a means of addressing this problem. Nonetheless, most scholars who conduct research about gifted children have failed to take the problem of the low ceilingsof avaik ble achievement tests into account. The teststhey use often cannot sample the full range of achievement amonghigh-IQ students. As a resul:, much existing research probably misrepresents the achievementof gifted students. Again, because of the inadequacy of availa'Ae tests, therehas been little alternative to such misrepresentation. Statistical regression toward the mean further confoundsunder- standing of the relationship between IQ and achievement.Although several of the factors just discussed contribute to thiseffect, chance factors also make it likely that students who score high at onetesting will score somewhat lower in a subsequent testing. Even if high-IQstudents were retested with the IQ test on which they were identified as gifted, aslightly lower group mean often would beobserved in their second scores. In summary, most studies find high-IQstudents' achievement test scores to be superior, butcloser to the mean than their IQ test scores. This fact is partly a result of differences in theskills and information measured by IQ and achievement tests. The presence ofconfounding variables, however, such as differences in testformat, low achievement test ceilings, and regression toward the mean suggests that the actualachievement level of many high-IQ students maybe closer to their IQ level than existing research shows.

Grades.In the case of high-IQ students, gradesmisrepresent academic progress even more than achievement test scores.In many instances, students' grades have little xo do with what theyknow about the subject on which they are graded (Chansky, 1964; Kubiszyn& Borich, 1984). Leiter and Brown (1985) found that primaryschool children's mathematics grades showed a slight correlation with theirdemonstrated mastery on standardized mathematics tests; but their readinggrades showed no correlation with their reading achievement test scores! lt can be arguedindeed, many teachersbelieve itthat such discrepancies indicate problems with achievement tests, notwith grades. This opinion is not surprising. Despite teachers' faith ingrading, however, many academically irrelevant factors enter into theirevaluation of students' performance. Students' physical appearance(Ysseldyke & Algozzine, 1982), their personality (Roedell, Jackton, &Robinson, 1980), race (Leacock, 1969), socioeconomic status (Rist, 1970), and sex(Carter, 1952) can and do influence teachers' judgments about studentsschool performance. This observation is not a condemnation of teachers,who cannot be blind to students' characteristics. It does suggestproblems with grades as a criterion of achievement.

1;9 hotelligesee Constructs 61 Teachers evaluate positively those children who are compliant and participate enthusiastically in classroom activities (Roedell et aL, 1980). These traits appear to be more prevalent in girls than in boys; a circumstance which may help to explain why teachers award girls higher grades than their performance warrants (Carter, 1952; Edmiston, 1943; Stockard, Lang, & Wood, 1985). The prevalence of these traits in girls may help account for the fact that about 60 percent of honor students arefemale whereas about 60 percent of students who make D or F averages are male. It may also help explain why boys are about twice as likely as girls to score higher on aptitude tests than on teacher ratings (Sexton, 1969). Compliance is also more prevalent in middle-SES than in low-SES students, and it is more prevalent in anglo students than in students from ethnic and racial minorities. Teachers' reactions to gifted students may depend, in part, on the students' race and SFS. In some classrooms of low- income students, teachers take a more negative view of &right black students than black students of average or low intelligence (Leacock, 1969). Rubovits and Maehr (1973) found that teachers praised white gifted children more than other children, but criticized gifted black children more than any of the others. High-IQ students who are independent and especially interested in theoretical and aesthetic issues sometimes make lower grades than other gifted students with lower IQ scores (Drews, 1961, cited in Tannenbaum, 1983). Teachers often consider bright but independent students to be less bright, less creative, lower in social acceptance, and more dogmatic than other gifted children. In a comparison of several indices of superior academic ability, Peg- nato and Birch (1959) found that grades did not discriminatewell between high-IQ students and students with average or slightly above-average IQ scores. When teachers used a 3.0 grade point average(GPA) as the criterion for referrals for the gifted program, mostof the students who were referred, about 86 percent, scored lower than 136 IQ, the minimum score for identification as gifted in the study. Among all of the children in the school who did score 136 IQ or higher, a large minorityabout 25 percenthad a GPA lower than 3.0. Discrepancies between IQ scores and grades are not due solely to the technical inadequacy of grades. Mediocre instruction, a slow pace, and academically irrelevant curricula may contribute to gifted children's underachievement. In general, when support for academic effort is not provided by the community, school, or home, achievement suffers. Of course, negative influences can also lower gifted children's achievement test scores; their effect on grades is more pronounced, however. Grades are more sensitive to compliance issues; and gifted children, like other children, sometimes refuse to conform to academic requirements. They may fail to turn in assignments, or they maymake careless

70 62 boreirsce Cosstructs errors. They may wander around theclassroom, daydream, or read rather than do their class work. If, as frequently happens, thesegifted children nevertheless learn the academic concepts and skills that aremeasured by achievement tests, their teachers are likely to regardthe test scores as erroneous.

Learning Rate High-IQ children's early mastery of academicsubjects is well documented. The reasons for their early mastery may vary,but they include quality of instruction, motivation, and effort; nodoubt an important con- tributor to learning rate is the effect that accumulatedknowledge has on the speed with which students can comprehend newmaterial. Although there are "late bloomers," these are the exceptionrather than the rule. Most research that describes gifted students' early masteryof auidem- ic skills is summative. It tells how advanced theirskills are at various age or grade levels. A smaller body of research isdet oted to the speed with which gifted students perform academic tasks.

Research on early mastery of academic shills.The results of a statewide study reported by Martinson (1972) typify thefindings about high-I(.; students' mastery of school subjects (e.g.,Gallagher & Crowder, 1957; Terman & Oden, 1947; Winy, 1930). Based oncomparisons of students' performance on standardized achievementbatteries, the report shows gifted children to be much further advanced inachievement than other children who have been in school an equal lengthof time. Gifted kindergarteners scored as high assecond graders on reading and mathematics tests. Gifted fourthgraders and fifth graders scored higher than seventh graders: in fact, the upperquartile of gifted fifth graders scored above the tenth-grade level.Most of the gifted eighth graders scored at the twelfth-grade level,and a sizable percentage scored as high as the average college freshman. Most ofthe gifted tenth graders and eleventh graders scored as high as average collegesophomores. A group of gifted high school seniors who were given theGraduate Record Examina- tion (GRE) in social sciences, humanities,and natural sciences scored as high as average college seniors. In the socialsciences, the gifted high school students' average was higher than collegeseniors who were social sciew majors (Martinson, 1912). The higher their IQ scores, the morerapidly children master academic skills. Hollingworth and Cobb's(1928) comparison of two groups of third graders. one with IQ scores rangingfrom 130 to 150 and one with IQ scores above 150, foundthat the children with higher IQ scores mastered third-grade skills faster. Case studies,biographies, and autobiographies of individuals with estimated IQ scores of 180and above (e.g., Grost, 1970;

71 Intelijence Constructs 63

Hollingworth, 1942; Wallace, 1986) supply anecdotal evidence that highly gifted children can, at the age of nine or ten, easily compete with average college students. Research on speed of performance on academk tasks.Whether or not gifted students complete individual academic tasks more rapidly than other children and the degree to which the speed of performance contributes to early mastery has not been established. In fact, the question has not even been studied very extensively. Speed of performance on academic tasks, in itself, may be too trivial to be a useful predictor of academic achievement. One study (Duncan, 1969) purports to show a positive correlation between the speed of accomplishment of academic tasks and IQ. Duncan compared the oral reading rates, verbal fluency rates, and arithmetic calculation rates of gifted fourth-, fifth-, and sixth-grade children with those of other students in the intermediate grades. She found that, on the average, gifted pupils had higher performance rates than other pupils. The greatest difference in performance rate on an academic task was in arithmetic calculations. The gifted sixth graders were able to perform correctly 28.4 multiplication operations per minute while the average sixth graders calculated correctly 17.5 operations per minute. The sixth-grade gifted students differed moie from the norm for their grade than did the gifted fourth- and fifth-grade students. A problem with Duncan's study is that the scoring procedures fac- tored accuracy into the rate score; the reported rates are rates of correct responses. Even if the gifted children's accuracy was greater than that of other children, it is still possible that they worked no faster. Some of' the materials used by Duncan were above grade level, so this explanation is likely. In fact, an earlier study (Klausmeier & Loughlin, 1961), which did not confound accuracy and rate, discovered no significant difference in problem-solving rate among high-, average-, and low-1Q students.

Family Support Parents of high-IQ students assist and guide their children's scholastic efforts. Because parental support is a consistently positive correlate of superior school performance, the nature and extent of the support are important intles in gifted education. Since family socioeconomic status (SES) influences the way parents bring up their children (Jencks et al., 1979), SES and related variables may help explain superior performance. Conversely, these variables may also help account for the failure of some high-IQ individuals to succeed in academic endeavors. Intellectual environment. The home environment of many high-IQ children is characterized by intellectual activity; reading; writing; listening to music; visiting libraries and museums; and attending movies, plays, and

7 2 64 woolly," Contracts concerts. Cornell (1984) found that, compared with other middle-class families, families with identified gifted children scored higheron an intellectual-cultural orientation scale designed to measure the extent to whicha family engages in political, social, intellectual, and cultural pursuits. Cornell also found that family "togetherness" was stressed to a striking degree in the families with gifted children. Parents of the gifted reported being closely involved with their children in family activities. Ter- man and Oden (1947) found that gifted children were more likely than other children to come from two-parent homes. Their finding may reflect the additional time and resources two parents can commit to their children's intellectual development, or it may reflect the high SES of their group. At any rate, the incidence of divorce and separation was lower in this group of parents than in the general population even in the 1920s. Anecdotal evidence also supports the positive influence of families on gifted children. Rhodes Scholars describe their families as strong and supportive. Their parents report that they were involved in their children's schooling and encouraged them to assume difficult tasks. Reading and books were frequently mentioned as part of the families' shared recreation (Denbow, 1984). Similar findings are reported for MacArthur Fellows (Cox et al., 1985). In retrospect, these high-achieving adults attribute their success to their parents' support of their intellectual development. They report that commitment to learning, to literature, and to the discussion of ideas was an important part of their childhoods. Although their educational levels varied, the parents shared an exceptional commitment to Fport- ing their children's intellectual efforts, both in and out of school.

Socioeconomic status and family support.Family background accounts for about half the variance in children's educational attainment; and economic status is a major determinant of the family's overall impact Uencks et al., 1979). The odds are against low-income families having the ability to offer the intellectual and financial support that can be provided by parents with high education levels and adequate incomes. Identified gifted children art predominantly members of middle- and upper-class families. Only 18 percent of the children identified as gifted in Ternian's study were the children of laborers (Terman & Oden, 1947,. Of those, only a few were the children of semiskilled or unskilled laborers. Of the gifted children who came from low-income families, a large proportion were underachievers. In 1947, Terman and Oden determined that on average, parents of children in the gifted group ha'i -ompleted four to five more years of formal schooling than the average person of the same generation in the United States at that time. The parents of the underachievers, however, averaged far below the achievers' parents in education; half of the most

7 3 Intelligence Constructs 65 successful group's fathers were graduated from college as compared to only 15.5 percent of the fathers of the least successful men. The occupational status of the underachievers' fathers was also markedlylower. Other family variables may relate to SES. These variablesinclude intactness of the family unit, mothers' employment, family religion, race, and ethnicity. Research that relates the absence of a father tochildren's cognitive performance is mixed. Some studies have found no difference in school achievement as a function of the father's absence from the home (Bronfenbrenner & Crouter, 1982); but a recent review of the literature found small, consistent differences favoring children from two-parent families (Hetherington, Camara, & Featherman, 1981). Bachman (1970) found slight detrimental effects on boys' verbal IQ when thefather was absent because of divorce or separation, but no effect whenthe father had died. The incidence of separation and divorce was about twice ashigh for parents of Terman and Oden's (1959)"least successful" men as for the "most successful" group. The proportion of deceased fathers wasalso considerably larger for the underachieving group. Mothers' employment, whether in a single-parent or two-parent household, does not have consistent or large effects on children's achievement either. Related factors, such asincome, however, do affect achievement (Hayes & Kamerman, 1983). Somestudies find that mothers of gifted children are more likely to be employed than those of otherchildren, particularly J the mothers are themselves gifted (Grath, 1975);but among most white, middle-class families,mothers' employment has little effec; on children's achievement. Among black, single-parent families, however, employment of the mother is positively correlated with achievement(Milne, Myers, Rosenthal, & Ginsberg, 1986). The family's ethnic, racial, and religious group membership mayalso influence children's school performance. In the United States, Jewishand Oriental families seem to value their children's educationalachievement most. In 1986, Asian-Americanstudents, who comprised only 2.1 percent of the college population, constituted 11 percent of thefreshman class at Harvard and 18 percent of the freshman class at theMassachusetts In- stitute of Technology (Butterfield, 1986). Jewish children also arestrongly represented in select colleges as well as in gifted programs.About twice as many Jewish children as would beexpected based on California's population at the time were identified as gifted by Terman in 1925.Students from other ethnic (e.g., Mexican-American) or racial (e.g.,American Indian) backgrounds, on the other hand, are underrepresented ingifted programs. Catholic children, as a group, score at the mean of IQ tests;Baptist children score the farthest below the mean of any religious group(Bach- man, 1970). These effects, likethose of ethnicity, can be attributed in part to SES. The family's economic status does not,however, explain all of the

74 66 Intelligence Constructs differences between Jewish, Catholic, and Baptistchildren's mean IQ scores; nor does it explainall of the difference between Asian-American and other immigrant children's scores. Educationalattitudes and aspirations vary to some degree with the groups' differentcultural histories, and these or similar differences may influence schoolperformance beyond the influence of SES. An example of a difference in attitude that issignificant for their children's intellectual development was found in astudy of mothers' views about academic achievement (New York Times,1986). Japanese mothers were most strongly convincedthat academic achievement was the resultof hard work; Chinese mothers were nearly asconvinced; but American mothers usually attributed academic achievement toinborn ability. The mean 1(2 of black students is usuallyreported to be 10 to 15 points lower than the mean IQ of caucasianstudents (Bloom, 1964). This difference is often construed to represent geneticdifferences in intelligence between the races.However, Bachman (1970) found thatdifferences between whites and blacks disappeared when heexcluded from his analysis black students from segregated southern schools.When whites and blacks from integrated schools were compared, theblack students' mean IQ score was only 3.3 points lower thanthe white students' mean IQ. Thisdiffer- ence is about the same asthe 3.4 point difference found betweenwhites in the Northeast and whites in the West when SFS wascontrolled. Even these slightly lower scores may be a function of SES. Bachman(1970) comments:

Even though we have invested much in our measurementof SEL [socioeconomic level), we surely are not completely successfulin our attempts to control it statistically. Moreover, we cannot saythat the black students in integrated schools have received "equal" treatmentthroughout their school experience. Some spent their grade school years insegregated schools: and some spent their high school yearsin course programs that are largely segregated. In short, statistical controls for SEL andschool experience are at best only approximations: and because of this, we cannotconclude that even the small difference of 3.3 QT would remain if otherfactors were fully and completely controlled. (p. 83)

Nearly every family variable shown to affectachievement is related to SES. When SES is controlled, differencesbetween the mean IQ scores of minority groups and white, middle-classchildren are practically eliminated. The effects of the family's SFS onchildren's school performance appear to be cumulative.Durkin (1966) found that socioeconomic status was not as influentialin early reading achievement as in laterreading achievement. Early reading related more to themothers' willingness and ability to involve their children with books thanit related to SES. Although many studies havefound that parental involvement and supporthas a

75 hatiligesure Consbucts 67 positive effect on low-income students' schoolachievement, these studies are often looking atdifferences within a particular SES group (e.g.,Dolan, 1983). Bloom (1964) found that at Grade 2, there was nocorrelation between reading comprehension andthe father's occupational level; atGrade 8, the correlation was 0.50. He attributedthis increase to the accumulated effects of socioeconomic status on schoolperformance. These effects include not only the nature and extent of the family'sinfluence on children's educational efforts, but also the nature and extentof the teacher's influence on children's educationalefforts (cf. Anyon, 1980).

Attainment Childpm who have high IQ scores when they areeight years old or older can be expected to have high IQ scoreswhen they are adults (Bloom, 1964). This phenomenon is partly related totheir longer period of school attendance. Although being smart does not makethem rich, it does predict that they will be more than ordinarilyproductive in academic and professional fields. Their above-average academicperformance, in many instances, continues throughoutadulthood.

Adult IQ of high-IQ children.In order to determine the stability of high IQ scores, Terman developed averbal intelligence test with items difficult enough to assess the vocabularyand general information acquired by his highly able adult subjects. TheTerrnan Concept Mastery Test (CMT) is a multiple-choice test of' the ability torecognize synonyms and antonyms and to recognizeverbal analogies (Terman, 1973). The mean CMT score of the group ofgifted subjects in Terman and Oden's (1947) study was 137; and the median was141. The mean score of subjects with 170 and above childhood IQ scores washighest: 155.8 on the CMT. Those with childhood IQ scoresof 135-139 had a mean of 114.2. Schooling also related to the subjects' CMT scores.Those who had doctorates scored a mean of 159.0; thosewho had medical degrees scored a mean of 143.6; thosewith master's degrees, 143; and thosewith bachelor's degrees, 135.7. These scores are muchhigher than the scores of other groups who took the test.The mean of graduate students at theUniversity of California, most of whom werecompleting their doctorates or medical degrees at the time they took the test, was118. Engineers and scientists in a Navy electronks laboratory, all collegegraduates, including several doctorates, had a mean of94.

Scholarship.Terman and Oden (1959) report in TheGifted Grim p at Mid-Lift that the high-IQ students intheir study tended to stay in school longer than others. About ten times as manygifted students in their study 76 68 hakiligesce Constructs were college graduates as were unselectedstudents. About 14 percent of the men and 4 percent of the women completed doctoral programs at a time when less than 3 percent of the population completed them. Among the subjects with 180 IQ or above, about half received academichonors, such as membership in Phi Beta Kappa or Sigma Xi (Feldman, 1984).Of Feldman's (1984) sample of the 140 IQ group, about one-quarter were awarded such honors. There were exceptions to the rule, however. The 150 least success- fur men in *rerman's study were less likely to attend college, and thosewho did mzde poorer grades than the most successful 150 menin the study. Twenty-five percent of the most successful group had an "A" average while only 5.8 percent of the least successful group had an "A" average.Only 7 percent of the most successful averaged "C" orlower, compared with 30.8 percent of the least successful group.Half of the most successful group were graduated with honors;only 14 percent of the underachievers were. This section of the chapter has reviewed the relationship between high-1Q giftedness and school perfonvance. A more complexview of academic expectations is presented in the next section. The expectationsthat result from an emphasis on the development of intellectimplicate certain school practices. 'Mese recommended practices,together with typical school provisions for the intellectually gifted, are consideredin the final section of the chapter.

INTELLECT

As implied in earlier sections of this chapter,definitions of intelligence are "culture-bound" (see e.g., Gallagher, 1975), because they are defined by cultural values. Very little, however, has been done to examinethose values and traditions that have determined our particularculture's notions of intelligence. These culture-bound notions are, essentially, historicalphenomena, and this section briefly examines the historical development ofthe institutions, values, and ideas that give meaning to the construct,"intelligence." Through this analysis, the discussion contrasts the notion ofintelligence with the notion of intellect. It applies these two notions tothe discussions of (1) the role and nature of scholarship and (2)the relationship between progress and thought.

Western Cultural Tradition Two trends in the Westert cultural tradition haveinfluenced our conception of intelligence: rationalism and literacy.Rationalism is reliance on a way of thinking (reason)that emphasizes the abstraction of general

7 7 Intelligence Cowbirds 69 laws from experience. It relies not on a framework offate or deity, but on a framework of purely human autonomy. Sense-data andempirical methods are, in this sense, tools of reason, not reasonitself. Literacy is familiarity with the written word. In addition to denoting the iibihty toread, it also connotes a sense of form and content inthe expression and comprehension of ideas. Literacy and reason constitute the intellectual substanceof scholastic aptitude, and they play a prominent part in theidentification and education of gifted children in an industrial culture. Thefollowing brief discussion of history is intended to suggest the ways inwhich social forces shape our work as educators.

The rise of literary and reason.Literacy and reason were not always accessible to many people. In ancient Athens, forexample, literacy was the prerogative of citizens, but only 10 percent of the totalpopulation of the Athenian state were citizens (Butts, 1973). The extentof literacy in the Roman empire is unknown (Bowen, 1972),but it was probably quite narrow. Between the fall of Rome and the European Renaissance, a greatdeal happened that would help literacy spread. As citiesbecame established in barbarian Europe, universities emerged as centers offormal learning, and trade also increased. Increased trade brought theWest knowledge and products that have helped shape the world as we know it.From North Africa came algebrathe foundation of modernmathematicsand from China an important product for the future ofliteracypaper. Literacy was strongly influenced, too, by the religious reform movementthat swept notthern Europe and caused major changes in the statusof the Church of Rome. Among the reforms proposed was anidea that was quite radical for its time: the idea that men and women shouldbe free to read the bible in their own languages. Bibles appeared for the first time in German.English, and other "vernacular tongues." Sometimes the translators paidfor their heretical work with their lives, so great was the fear ofliteracy. Although the advent of paper, printing, and the publication ofbibles in languages other than Latin helped extend literacy, still, in alllikelihood, less than one-half of 1 percent of the general population (I personin 200) was literate in 1700. Modern readers, of course, need to remindthemselves that, at that time, reading was not much of a practical advantage inlife. The Renaissance established the relevance ofhuman beings as a fit object of study (humanism), but the Enlightenmentthat followed firmly established humanity, rather than God orChurch, as the prima rv subject of this world. The West was on the verge of a newliteracya new sense of its power in the world and a new senseof the role of thoughtof observation, of reflection, and of imaginationin that world.

7 8 70 lidelligswee Constructs

By 1800 basic literacy was becoming an established phenomenon even outside governments, churches, and universities. Still, the absolute number of readers was surprisingly small even in 1800, no more than 2 percent of the total population (Watt, 1957). This is a fact we tend to forget in an era when nearly everyone can read a little. (Of course, reading only a little is today a much greater handicap than not reading at all was in 1800!)

The Role and Nature of Scholarship Educators should understand the role of literacy and reason in West- ern civilization, because, sustained partly through thescholarship of intellectuals, literacy and reason contribute significantly to the making of new meaning and the stewardship of wisdom. Scholarship, however, involves more than intelligence; it involves intellect. Richard Hofstadter (1963), a noted historian, distinguished between intellect and intelligence. Hofstadter saw intellect as"critical, creative, and contemplative," and intelligence as "manipulative, adjustive, and unfailingly practicai." Hofstadter (1963) looked at how our society views intellect and intelligence when he wrote:

The man of imelligence is always praised; the man of intellect is sometimes also praised, but he is often looked upon with resentment or suspicion. It is he, and not the intelligent man, who may be called unreliable, superfluous, immoral, or subversive; sometimes he is even said to be, for all of his intellect, unintelligent. (p. 24)

Hotkadter's remarks help us see why intelligence is not always used to enhance intellect. Schools may have a vested interest in making sure that high-IQ students in the end put their talents to work in enterprises where intelligence can make an immediate, practical contribution. Scholars, after all, have an unwelcome habit of asking embarrassing questions.

Scholarship and scholastic aptitude.More than native intelligence. scholastic aptitude is what IQ tests measure. Scholastic aptitude is the capacity for scholarship. A child who arrives at school with a "high IQ" has an internalized ability to do academic tasks well. Schools, however, may ignore this ability. Instead of promoting habits of scholarship and, hence, involvement with literacy and reason, schools may instead promote habits of cieference. or- derliness, self-restraint, and regular attendance. They may bend studems' talents to the application of socially acceptable work that promises reasonable practical benefit. The emphasis placed on mathematical and scientific training for gifted students is in some ways a good example of this tendency (cf. Tannenbaum, 1981,1983).

79 Intelligence Construes 71

In this way schools may, in fact, damage thetalents of their most academically able students. Schools may also fail to recognize thescholastic aptitude of many studentsespecially those in ethnic,racial, or regional minoritieswho do not perform well on IQ tests.These students may nonetheless be demonstrably apt, even though their aptitudehas not been bent to academic tasks. By characterizing these students asunintelligent, schools may exclude them from any experiencesthat would enhance their intellect. Stereotyping of these students' educability isusual, and it may, in fact, serve to deprive them of access to literacyand reason.

Progress and thought.Certain political and economic premises of our society depend onthe belief in human excellence. This belief supports a meritocratic view of thedistribution of wealth and power (see Chapter 2). Even though this view is not based on an accuraterepresentation of our political economy, it nevertheless explains thepropensity to see intellectuals as symbols of human excellence. As a society,however, we respond defensively toward intellectuals; we seem tolike them a lot better once they are dead. After their deaths, we aresafe from their personal influence, their embarrassing questions, and theirimmediate example all of which make us uncomfortableand we cansafely regard them as symbols. But society also regards intellectuals as dangerous.Intellectuals challenge the status quo. They are critical of society, andthey are hard to control. Intellectuals also tend to concern themselveswith issues that do not relate directly to the practical endeavors of our economy.which are seen as the development and the distribution of goodsand services that have the potential to make money. As ,. society, we put a lot more of our resources intothe promotion of business and government than we do into thepromotion of literacy and rational thought. Many Americans see progress only in termsof economic and technological advancement. They do notregard ideas as significant contributions to the progress of our culture. Nevertheless, the intellectual legacy is very important to thefuture of our society.Western civilization may need a vital cultural legacyin order to save itself. We may need somepowerful ideas to help us resist or recover from the destructiveness and alienation producedby unrestrained technological development (Leontief, 1982; Read, 1960;Thurow, 1987). I n the long run, progress may depend more oncareful thought than on the manufacture of tangible and marketable products. Intellectually gifted students have demonstrable aptitudefor scholarly work, and programs for them shouldaddress this aptitude, above all else. The section that follows provides abasis for understanding the current status of such programs and forevaluating their relevance.

0 72 biteiligenceCoign:as

SCHOOL RESPONSES TO HIGH-1Q CHILDREN

Of the constructs of giftedness, IQ is the easiest to measure. It is also the construct that is most closely related to the academicrequirements of school. Consequently, programs for high-IQ students were instituted earliest and continue to be most prevalent. This section of the chapter first provides an historical review of programs for high-1Q students. Next, it recommends educational approaches that involve such students in the intellectual traditions of literacy and reason.

Historical Overview Accelerution was probably the earliest school response to intellectually gifted students. In one-room schools, acceleration took place when a child mastered the curriculum of a particular grade level. The advent of graded schools, however, made acceleration seem less natural. Nevertheless, grade skipping took place often. Some districts conducted graded schools in Let fits of one-half year. Students could,therefore, skip half a grade at a time. In the early twentieth ccntury, educators began to organizespecial classes for high-1Q children. Leta Hollingworth helped establish one of the earliest classes for such children. Public School 500 (also known as the Speyer School) in New York City had special classrooms for both giftedand highly gifted children. These classes combined acceleration and enrichment (Hollingworth, 1926). They also provided a moreprogressive curriculum fiir the gifted than the curriculum 0. ''..!red in that era to average students. Other classrooms were established in major cities such as Cincin- nati, Cleveland, and Los Angeles. The Major Works Program, which started in Cleveland in 1921, exemplifies these efforts.

The influence of progressive education.Enrichment, as an alternative to acceleration, resulted from the concerns ofprogressive educators (e.g., Osburn & Rohan, 1931). Progressive educators were interested in more than !he academic development of children; they were interested inthe development of the "whole" child. 'These educators viewed enrichment as a means to allow high-1Q students todevelop appropriate academic skills without having to leave their age group peers (Fowlkes, 1930). This view dominated gifted education from the 1930s through the 1950s; itbecame popular again in the late 1960s and continues to influence programsfor high-IQ students. Not only did the aims of enrichment programs result f"rom the notions of progressive educators, the content of such programs alsoreflected progressive beliefs. Far example, interdisciplinary instruction wasimple- mented by progressive educaters (Cremin, 1961) and continues tocharac-

81 Intelligence Coutrires 73 terize enrichment programs (see e.g., Cox et al., 1985; Martinson, 1968; Renzulli, 1977). Progressive views of education also influenced other characteristics of programs for high-IQ students. Programs to develop critical and creative thinking skills were recommended by pmgressive educators (e.g.. Sted- man, 1924). Progressive educators also noted the importanceof individual- ization, independent study, consideration of real problems, student-directed learning, and out-of-school learning for highrability children. These instructional approaches are still prominent among contemporary recommendations for intellectually gifted students.

The influence of Spinik. The 1957 launching of the Russian satel- lite, Sputnik, altered U.S. priorities for the education of intellectually gifted students (Tannenbaurn, 1981). In the late 1950s and the 1960s, educators stressed instruction in science and math. This priority prompted the development of innovative and challenging curricula (Gallagher, 1975).The earlier emphasis on enrichment yielded to a new focus. To organize the efficient delivery of difficult curricula in science and math, schools began increasingly to group students by ability or achievement levels.

Equal opportunity and back ta basics.In the mid- and late 1960s educational programs were guided by the concern for equity. Programs for disadvantaged children proliferated. Programs fin high-IQ children were seldom funded. Nevertheless, many intellectually gifted students continued to receive relevant academic instruction. Many schools continued to use the innovative curricula developedin the late 1950s and early I 960s. The theoretical orientation of "new math," for example. challenged bright students even where special programs were not provided. In addition. schools persisted in grouping children by ability until law suits in the early 1970s discouraged this practice. In response to the concern for cultural and individual self-determina- i'on that dominated educational thinking in the mid- to late 1960s, educators in the mid- 1970s and 1980s showed a renewed concernfor "basic skills." Whereas instruction in basic skills is intended to help more students pass minimum competency tests, it fails to teachliteracy and reason. "Basic skills." unfortunately, do not include the habits of critical thinking and inquiry. The educational legacy of the last 30 years has influenced the most current recommendations for theeducation of intellectually gifted students (e.g., Bennett, 1986; Cox et al., 1985; Stanley, 1981). These recommendations advocate:

1. acceleration. 2.rigorous coursework and emphasison academics.

2 74 hitgaigraes Conantets

3.foreign language instruction. 4.cultural literacy, 5.high expectations. and 6.advanced instruction in math and science.

Contemporary Practices Programs for intellectually gifted students have not alwaysaddressed the cognitive strengths that these students demonstrate.If educators choose to design programs on the basis of the recommendationsabove, however, they will be more likely to develop these strengths. To putthese recommendations into practice, educators will need to establishpractical routkies of identification, program planning, andinstruction. The discussion below describes practices that will be most likely tocorrespond to the academic needs of intellectually gifted students. Thesepractices should make it possible for high-IQ students to cultivatehabits of scholarship and, thereby, to understand and, perhaps more important, todevelop our intellectual legacy still further.

Identification.Intellectual giftedness is most often determined on the basis of1(2test scores. Typically, studentsqualify for programs on the basis of scores that are at least 2 standard deviationsabove the mean on tests of intelligence. IQ tests measureintellectual potentialthat is, how well individuals can perform academic tasks. In an earlier part of the chapter we termedthis capacity "scholastic aptitude." Tests that measure scholastic aptitude are,therefore, also appropriate for determining intellectual giftedness. Measuring academic achievement is also a way todetermine intellectual giftedness. It is impossible for students to score2 standard deviations above the mean on valid and reliableachievement tests without having the potential to do so. If such a case arises, the accuracyof the student's IQ test score needs to bequestioned; it may be an unreliable measure of the student's giftedness. An important consideration inthe use of any of these sorts of tests is their fairness to children from minority groups.Both in Chapter 2 and in this chapter, we stated that students fromminority groups score lower than white, middk-class students on standardized tests.Because their lower scores reflect variablesother than intelligence, educators must use identification procedures that assure the equal accessof minority students to programs for theintellectually gifted. A promising approach to culture-fair assessme ainvolves the use of local, regional, or minority-group norms.Educators can use local or minor- ky-group minas to identify thechildren with the highest scholastic aptitude 07 achievement in a particularschool or from a particular subgroup of the populaticn. 83 hoelligence Canonici: 75

Another appioach is to modify the operational definition of giftedness to allow for the variability of scores among children from minority groups. The model definition given in Box 3-1 reflects the concern fol. equity. It also bases eligibility on either an intelligence or an achievement test Sane.

BOX 3-1 A Model Definition

Except in :he case of a nonadvantaged student, a student shall be declared eligible to receive special services under the auspices of gifted education if either of the following conditions obtains: (1) the student scores two standard deviations above the mean on any major portion of a comprehensive evaluaeon-level test of achievement (e.g., the mathematics cluster of the Woodcock-Johnson Psychoeducational Battery, Part 11 Tests of Achievement) in consideration of one standard error of measurement, OR (2) the student scores two standard deviations above the mean on a comprehensive evaluation-level test of intelligence (e.g., the Stanford-Binet Intelligence Scale) in consideration of one standard error of measurement. /n the case of nonadvantaged students, a declaration of eligibility shall occur in consideration of three standard errors of measurement. The term "nctladvantaged" refers to ethnically different students, to students for whom English is not the native tongue, and to students who are eligible to receive free or reduced-price meals. (Howley, 1986. p. 23)

The most appropriate operational definitions of intellectual giftedness depend on measurement. Although there are manyinstruments that purport to measure intelligence or achievement, the most %alid and reliable ones are comprehensive and are administered 'ndividually. These tests are classified as evaluation instruments. They are contrasted with screening instruments, which are useful only fbr identifying children who might score at a particular level on an evaluation test. Evaluationinstruments are the only hinds of tests that should be used to determine eligibility for placement in a special program. Two scales of in.Jligence are most often used to determine intellectual giftedness. They are the Stanford-Binet Scale and the Wechsler Scale. The Stanford-Binet Scale is based on a definition of intelligence that equates IQ with scholastic aptitude. For this reason, the Stanford-Binet scale includes a preponderance of verbal items (Sattler, 1982). It includes items that measure vocabulary, abstract words, sentence building, similarities and differences, analogies, sentence completion, verbal absurdities, and reasoning. By contrast, the Wechsler Scale is based on the view that intelligence is the global ability to solve problems. It samples two broad domains of cognitive functioning, the verbal domain and the performance domain. There are many other intelligence scales that are useful for special populations. The scale developed by the Kaufmans was intended to help identify intellectually gifted students from culturally different back-

4 76 hstellOrsce army:acts grounds. Unfortunately, the K-ABC has not been assuccessful at this endeavor as was first hoped. Evaluators can also find a number of tests that measure scholastic aptitude. Most of these tests, however, are designed for groupadministra- tion. The aptitude test that is most often used toidentify intellectually gifted students is the College Entrance Examination Board's(Educational Testing Service, 1926-1987) Scholastic Aptitude Test (SAT).SAT scores are used by selective colleges (manycolleges are not selective) to make decisions about admissions. They are also used fordetermining the eligibility of younger students (seventh andeighth graders) for special programs (Stanley, 1977). Individual tests of achievement, unfortunately, often do nothave a high enough ceiling to measure reliably the achievementof intellectually gifted adolescents. With younger children, evaluators can,however, use one of several tests. Some of these tests arebatteries that measure achievement in a number of different areas;others measure achievement in only one subject. Most prominent amongthe multiple-subject batteries is the Woodcock-Johnson Psychoeducational Battery (Woodcock &Johnson, 1977). This test includes measures of achievement in four areas:mathematics, reading, written language, and knowledge.

Pnigram pia/ming.Regardless of whether an individual teacher or a committee plans the instructional program for anintellectually gifted student, certain principler should be considered.First, the program should address the student's demonstrated propensity foracademic learning. Sec- ond, the program should guide decisions aboutplacement. The student should be placed in environments where his orher needs can best be met. The gifted resource room often is not the optimalenvironment for a student who exhibits advanced skills in academicsubjects. Finally, the program should account fbrlong-range planning. A student who is placed in advanced classes in elementary school should not have to repeatsimilar classes at a later point in his or her education. The availability of a number of program optionsfacilitates planning fbr intellectually gifted students. Unfortunately, mostschool districts endorse only one or two options (Cox et al., 1985). Thetwelve options listed in Box 3-2 describe the range of services requiredfor planning Rppropriate instructional programs.

BOX 3-2Effective Program Options

First Option (Elementary Level): Each elementaryschool provides special accelerated reading andmathematicsclasses for gifted primary students. Second Option (Elementary Level): Each elementary schoolprovides a review process to consider yearly, or more frequently upon request ofteachers, parents, or students, which bitellignice Constructs 77

accelerative strategy or strategies (e.g., cross-class placement, special accelerated classes, early entry, dual attendance, combined grades, grade skipping) are appropriate for each gifted student. Third Option (Elementary Level): Each elementary school offers a choke of three years of either of two foreign languages to gifted students. Fourth Option (Elementary Level): Eath elementary school offers an accelerated literature and writing class that prepares Ptudents to write good ten-page essays by age 13. (A good essay conforms to high standads of grammar, diction, and style. High standards for gifted students aged 13 shall approximate the best efforts of average first-year undergraduates.) Such a course will provide 1 Carnegie Credit in high-school English. Fifth Option (Elementary Level): Each elementary school provides an advanced mathematics class to deliver the complete Algebra I content (linear and quadratic equations through the derivation and appli-ation of the quadratic formula) to gifted students in Grades 4-8. Such a course will pmvide 1 Carnegie Credit in high-school mathematics. Sixth Option (Elementary Level): Each elementary school provides an accelerated science class in biology or chemistry that will provide 1 Carnegie Credit in high-school science. Seventh Option (Secondary Level): Each secondary school develops procedures that allow students to earn course credit by examination (initially in a limited number of courses). Eighth Option (Secondary Level): Each secondary school develops an accelerated (3 or 4 years in 2 or 3 years) mathematics sequence that includes geometry, a second course in algebra, trigonometry, and integral and differential calculus to begin in the ninth grade. Such a sequence provides 3 or 4 Carnegie Credits in high-school mathematics. Ninth Option (Secondary Level): Each secondary school develops an accelerated (4 years in 3) literature program that requires the reading of a major work each month and frequent writing about works read. The course sequence should cover British and American literature, foreign literatuie in translation, and philosophy (using original sources). Such a sequence provides 4 Carnegie Credits in high-school English. Tenth Option (Secondary Level): Each secondary school develops a strategy to enable gifted students to complete the science sequence that provides 3 Carnegie Credits in biology. chemistry, and physics by the end of Grade 11. Eleventh Option (Secondary Level): Each secondary school develops an advisement system that ensures that an increasing number of gifted students attend very selective institutions of higher education. Twelfth Option (Secondary Level): in cooperation with other institutions each secondary school develops an internship program that gives interested gifted students a substantive experience with an outstanding community sponsor during the senior year of high school (for those gifted students who elect a four-year high school program). (Howley, 1986, pp. 24-25)

Appropriate instruction.One notable omission from the range of services listed above is the resource room. Although this method of deliver- ing special education is the most common for high-1Q children, it is the least effective (Cox, et al, 1985). Resource room programs most often provide enrichment, a kind of instruction that, by definition, is tangential to the content of the regular curriculum (Vernon, Adamson, & Vernon, 1977). There are two reasons why enrichment, as commonly practiced, is an inadequate alternative for high-IQ children; (1) enrichment activities lack substance and (2) other options are much more relevant to the education of' high-IQ students (Stanley, 1981). Though some progressive educators of the gifted in the 1920s and 1930s had quite specific activities in mind, 78 htteiligesee Constructs enrichment today is often so broadly conceived that it is actuallysuitable for all children, regardless of IQ (Weiler, 1978). Like extracurricular activities, enrichment programs often do provide positive andenjoyable experiences. Special programs designed for high-IQ childrenought, however, to provide a lot more: They shouldprovide substantive instruction in academic subjects. Two instructional strategies are suited to the needs ofhigh-IQ children: acceleration and special curriculum. Accelerationallows intellectually gifted students to progress academically at a rate thatapproximates the rate at which they learn. Because acceleration is amodification of the regular curriculum, it is easy and inexpensive to use. A specialcurriculum is more expensive and more complicated to organize.This approach, which combines acceleration and ability grouping, providesthe only comprehensive way to address all of the academic needs of high-IQstudents. Such a curriculum can be delivered exclusively togifted children, or it can be offered to high achievers as well.

SUMMARY

This chapter viewed the psychological constructof intelligence from several perspectives. It examined theories ofintelligence including behavioral theories, single-factor theories, multiple-factortheories, dichotomous-ability theories, and information-processing theories.The chapter contrasted the more limited construct of scholasticaptitude with the broader construct of adaptive behavior. By reviewing the literature on the schoolperformance of high-IQ children, the chapter showed the relationship ofintelligence to indices of academic attainment. This section indicated that,in spite of some limitations, IQ tests can be used for predictingschool performance. Because of the strong correspondence betweenintelligence and scholastic aptitude, the chapter also emphasized the importanceof intellectual work for individuals with high Igs. Finally, the chapter provided a review ofschools' options for educating high-IQ children. It suggested adefinition of intellectual giftedness and suitable methods for measuring IQ.It also discussed program alternatives that seem to be most effective forhigh-1Q students.

8 7 Creativity Constructs

I. Focusing Questions II.Intelligence and Adult Productivity III.Issues of Convergence and Divergence A.Questions and Answers 1. Kind of question 2. Kind and degree of response 3. Objectivity and divergent production B. Historical Notions of Creativity 1. Useful novelty C. Productivity and Creativity D.Making Sense of Creativity IV. Personality Traits Associated With Creativity A.Early Studies of Creative Genius 1. Psychoanalytic accounts of creativity 2. Humanistic psychologists' view of the creative person B. Recent Research Findings on Creative Adults 1. Interest in abstract ideas 2. Lack of interest in socializing 3. Unhappy childhood experiences 4.Emotional intensity 5. Emotional health 6. Emulation of nonconformist behavior C. Personality Traits of Children Who Are Divergent Thinkers 1. Divergent thinkers in the classroom

79 BO Creativity Cometnacts

2.Interactions between IQ and divergent thinking scores 3.Children who resemble creative adults V.Creativity and Intelligence A. The Creativity Construct 1. Creativity as a noncognitive trait 2.Creativity as fluency, flexibility, originality, and elaboration 3.Creativity as remote association 4.Creativity as fluency and uniqueness B.The Relationship between Creativity and Intelligence 1,The low correlation between creativity and intelligence 2.The high correlation between creativity and intelligence 3.Measurement cautions 4.Intelligence as necessary but not sufficient for creativity C.Creative Potential and Intellectual Potential 1. Do creativity tests predict academic achievement? 2.Do creativity tests predict creative achievement in adulthood? VI.School Responses to Creativity A.Spiritual Schooling B.Creativity Training in the Schools VII.Summary

Focusing Questions

'1. Is it always possible to distinguish convergent from divergent questions? 2.What actual personality traits and personal values lie behind the stereotype of the wild and rebellious artist? 3. In what different ways has creativity been operationalized as a psychological construct? 4. Why do some researchers believe there is a low correlation between IQ and creativity, whereas others believe there is a high correlation? Which researchers do you think are right? Why? 5.Educators often express a commitment to meeting the intellectual, physical, social-emotional, and spiritual needs of students. How does creativity training serve this commitment?

INTELLIGENCE AND ADULT PRODUCTIVITY The investigation of eminent individuals in the nineteenth centuryhad an important influence on gifted education. Chief among the reasonsfor investigating eminence was the perceived need for more ofthe kinds of Creativity Couteuets 81 products developed by the eminent. The industrial world was emerging rapidly, and the need to cultivate talent in science, industry, and commerce was keenly felt. The early researchers asked why their eminentsubjects were able to produce more and better works than other people.They answered that their subje9 were more intelligent in general, more talented in their particular work, and that they worked harier than other people (cf. Gallon, 1869/1962). By 1910, psychology had emerged as a science, and both psychologists and educators felt that the time had come to find and to train exceptional talent*. Although Terman (1925) established that his test did, in fact, predict achievement in school, he was subsequently unable to prove that his test predicted adult eminence (Feldman, 1984). What had gone wrong? One answer that seemed likely to many psychologists andeducators had been apparent to most skeptics of the IQ all along: IQ tests fail to measureall the components of intellect. Another answer that seemedlikely was that IQ has little to do with real-life achievement.

ISSUES OF CONVERGENCE AND DIVERGENCE

According to Guilford (1950), for example. IQ tests measured convergent thinking,process that results in finding a correct answer to acertain type of problem. The tests, claimed Guilford, fail to measure thvergent thinking. a process that results in producing manypossible solutions for a certain type of proJem. Guilford's observation intrigued E.Paul Torrance, a war- time colleague. Torrance had alrrady been puzzling out the notion of creativity for some time (Torrance, 1984). Torrance (1984) reports that as early as 1943 he had developed a creativity test. Guilford's notion of divergent thinking influenced Tor- rance's construct of creativity, and in his creativity tests T'orrance usedfour subskills reputed by Guilford to characterize divergent thinking. These subskills are (1) elaboration, (2) flexibility, (3) fluency, and (4) originality (see Box 4-2). Not all interpretations of creativity have adoptedGuilford's model. His notion of divergent thinking is nonetheless the historical beginning of the empirical study of creativity.

Questions and Answers Despite their important differences, convergent and divergent thinking are similar because they both answer questions. The differences in con iergent and divergent thinking oftenlie in the kind of question answered. They always differ, howmr, in (1) the kind and the degreeof response that is acceptable, and (2) the kind ofproduct that may be implicit in the response. 82 Creativity Constructs

Kind of question.Convergent questions have an answer du), is logically correct or that some authority has determined to be correct in advance. An unambiguous example is "What is the square root of nine?" A more ambiguous example is "What is the chief cause of the Civil War?" followed by answers in a multiple-choice format. Divergent questions do not have a correct answer as determined by the logic of some discipline or some authority. Divergent questions areby nature ambiguous. In fact, a divergent question may berephrased in the pi-cress of being answered. For example, "Whereshall we go this after- noon?" may evoke the response, "Nowherelet's stay here and makeice cream!" or even, "I'm not going anywhere with you!" depending onthe circumstance. In the first case, the reformulated question is "What shall we do if wt. stay here?" and in the second, "Would you like to go somewhere with me?" Simple divergent answers often beg the question. Elaborateddiver- gent responses provide a more complete contextin which the reformulated question and the thinking behind it become explicit A more ambiguous example of a divergent question is "How can you trisect an angle with compass and ruler?" Though the judgmentof history has thus far been that this task is impossible, the correct answer maystill be some undiscovered set of instructions. Onthe other hand the answer may be "You can't." The latter response challenges the question andmight be either convergent or divergent. Notice that an unsound convergent question (i.e., one thatdoes not, indeed, have a correct answer) resembles a divergent question, andthat a predetermined divergent question (i.e., one that specifies results) resembles a convergent question. The critical distinctionbetween convergence and divergence seems to involve authority. When a question is treated convergently, both the personputting the question and the person answering it treat the question assound. The person answering the questionweisarily accepts the soundness of the question on the authority of the person asking the question. Theauthority of thr questioner is not nermarilv accepted when a question isdivergent. Many questions can easily be approached convergently ordivergently, depending On the respor -lent's view ofthe question.

Kind and degree of response.Simple convergent responses resemble the solutions to puules or riddles. "I-heyfit into a predetermined context. Convergent responses must also be interpreted as makingunambiguous sense in such a context.They are COTItice and analytical. Thus, responses are objectively evaluated as part of a respondent's convergentthinking. Simple divergent responses, however, resemble lists or puns.They reveal connections, commonalities, or contrasts. They at efluent and .syntIwt- ic. Rather than fitting into a predetermined context,simple divergent re-

91 Creativity Construets 83

BOX 4-1Contrastingand Comparing Convergent and Divergent Thinking

A DIVERGENT VIEW Many commentators see convergent and divergent thinking as opposites. Most of the adjectives below are taken from the discussion in this chapter.

Adjectives that may describe convergent Adjectives that may describe divergent thinking: thinking: objective subjective normative autonomous analytic synthetic concise elaborate hard-working playful

Lists of adjectives might be said to be a divergent description of divergent and convergent thinking. This description is merely suggestive; the original list from which these adjectives were drawn contained 15 items.

A CONVERGENT VIEW Divergent and convergent thinking can be distinguished by one sufficient feature:

The feature that distinguishes convergent from divergent thinking is submission to external authority and objective evaluation. Convergent thinking submits, whereas divergent thinking does not. By operating autonomously, divergent thinking acts subjectively, whereas by operating normatively, convergent thinking acts objectively.

This statement distinguishes between divergent and convergent thinking in a convergent mode. It purports to be logical and concise. spouses suggest a new,altered, or extended context. This suggestive fluency means that elaboration may be necessaryforacomplete response. Fluency seems, however, to be the minimum observable characteristic needed to distinguish a simple convergent from a simple divergent answer (cf. Hocevar, 1979, 1980; Perkins, 1981). Originality is also said to be another significant characteristic of divergent thinking (cf. Guilford,1959; Koestler,1964). Originality, however, is usually inferred from the novelty of divergent responses. This inference does not necessarily mean that allnovelresponses are divergent. Novel responsesran beproduced by convergent thinking, as we shall see. Truly original (not merely novel) responses cannot be produced by convergent thinking, however, because original responses cannot be submitted to objective evaluation during their formulation. Original respottses mustbe suertively evaluatedby the respondent during their production. The subjective quality of originality (as opposed to novelty) indicates again the degree to which divergent thinking proceeds onits own authority. Originality is in this senseautonomous.Marcuse (1978), for example.

9 2 84 Creativity Constructs believes that works of art portray an autonomous vision of the world inherent in artists' manipulation of aesthetic form.

Objectivity and divergent production.The one sense in which it is possible to attribute objectivity to a divergent response, however, is the case in which an object is made. The reation of an object may constitute a fully elaborated divergent response, if the object is evaluated only subjectively by its creator(s) during production. Observers can subsequently make a normative (i.e., objective) comparison of the object with other objects to deterraine its utility, novelty, or beauty. An inference about the divergence of the producer's thinking may be made from the object's utility, novelty, or beauty. Many useful, novel, or beautiful objects are, however, in no way creative, or even the product of divergent thinking (Perkins, 1981). In particular, it seems strange to expect original products to conform to conventional standards of utility, novelty, and beauty. Nonetheless, definitions of creativity typically expect original products to conform to such standards, and most creativity research specifies that utility and novelty constitute necessary characteristics of creative (not merely divergent) products (see, e.g., Keating, 1980a; LaChapelle, 1983; Pearlman, 1983; Weinstein 8c Bobko, 1980; but cf. Marcuse, 1978; Perkins, 1981). I. storical Notions of Creativity The preceding discussion gives us a contemporary vantage point from which to examine the act of creation. However, it is interesting to see how the concept of creativity has changed historically since thenineteenth century. During the first half of the twentieth century,it was commonly assumed that creativity was the province of comparatively few individuals, that these individuals were primarily artists, and that creativity was not necessarily a desirable attribute, at least from the standpoint ofobjective social norms (LaChapelle, 1983). By the mid-twentieth centu, y, the concept of creativity changed (LaChapelle,1983; Pearlman, 1983), and creativity was found to be a potential ability of all individuals,applicable to all fields of human endeavor (see, e.g., Rogers, 1959, p. 71). Most subsequent empirical work on creativity has, in fact,contributed to a broadened definition ofthe term. According to LaChapelle (1983, p. 132), the broadened definition "allows for the possibility thatalmost anyone under almost anycircumstance can be creative." The construct of creativity has thus come to resemble the construct of intelligence. One may be intelligent or act intelligently without ever significantlyengaging the intellect; similarly, it seems in the broadened definition that one maybe creative or act creatively without ever creating anything. Nonetheless, recent programs for developing creativity have applied the broadened notion of creativity especially toindustrial and scientific

93 Creativity Caastrsiets 85 productivity. Osborn (1953/, an advertising exec/give, devised techniques of "creative problem solving." His training program was intended to foster the inventiveness and productivity of industrial research and development teams. Osborn's thesis was "that the economic supremacy of our countryf is maintained] by the creative ability of our citizens" (Osborn, 1963, p. x). Together creativity and patriotism became the province of business and industry, much as high IQ and patriotism became the province of math and science education in the 1960s (cf. Tannenbaum, 1981).

Useful novelty.The change in focus from the arts to business and industry is quite dramatic. Is there a link between industry and art that might help account for this historical change in focus? One link may be the notion of originality, er novelty (me..sured empirically as uniqueness, but cf. Perkins, 1981, pp. 282-283 an 'elsh, 1975, pp. 3, 69, for different notions of originality). With the h of vigorous advertising, industry in- duces new markets to buy new products. Novelty represents a competitive advantage in the marketplace. Novelty also has a place in the arts, where originality, like beauty, is sometimes valued as an end in itself. Novelty is characteristic of artistic modernism (Read. 1959). It can help an art dealer establish a market by relating the works of new artists to the dominant aesthetic of the times. Mere novelty, however, is unlikely to be an adequate proxy for originality. The sort of novelty shared by commercial and artistic products is novelty in trade. This similarity does not establish the originality of an artistic or commercial product, nor does it establishcommercial and artistic production as equally creative fields.

Productivity and Lreativity Creativity research has devoted comparatively little attention to the creative product (Perkins, 1981). It is difficult to see why. After all, the presumed need for such products is the motivation for this research. Per- haps the evaluation of created objects is considered to be the territory of art critics and scholars alone. Instead of assessing products, much of the creativity research, at pm- ent, assesses divergent thinking. The mostserious flaw in such a choice is that divergent thinking is not synonymous with creativity. Some researchers are aware of the importanceof this distinction (e.g.. Hocevar, 1980; Klein, 1982; Perkins, 1981) but others are not (e.g., Torrance, 1963, 1984; Wal- lach & Kogan, 1965). Two serious problems are caused by a failure to consider the distinction between divergent thinking and creativity. First, the concept of divergent thinking does not account for the act ofmaking a product. Empirical research into divergent thinking struggles to characterize the internal 86 Creativity Constructs states that precede the behavior of making a product (e.g., Wallas, 1926). Second, failure to distinguish productivity from creativity occurs when researchers are unable (e.g., MacKinnon, 1962/1981) or theorists refuse (e.g., Rogers, 1959) to distinguish the qualities that characterize produas. Of the terms, creativity and productivity, "productivity" is the more general term. It refers to the provision of goods or services in abundance. Such goods and services are typically useful, especially in trade. Concern for productivity reflects the entrepreneur's objective of increasing profits. Osborn (1953) noted explicitly that the primary function of his creativity training program was to increase the profits of industry and commerce. This analysis shows why the term creativity was broadened to include business and to include being and doing as well as making. Businessincludes manufactures (making) and services (being and doing), and its chief concern is profit. As Perkins (1981) notes, however, even quite worthwhile products need not be creative. He sees the quality of products (kind and excellence) as determining what is creative. Making Sense of Creativity Recent theoretical work has recognized that both convergent and divergent thinking are components of creativity. The central importance of a "creative" product (whatever it may be) has alsobeen recently acknowledged (e.g., Pearlman, 1983; Perkins, 1981). Both Pearlman and Perkins, however, neglect the impi rtant concept of autonomy. The reasons for this neglect are inherent in the nature of ordinary commerce and scholarship. In commerce, practical restraints place severe limitations on autonomy. Osborn's (1953) version of creativity training respects such limitations in its group method. The group method maximizes the benefits of divergent thinking, while minimizing the risks of true autonomy (see e.g., Abelson, 1967). The groupmethod is a practical restraint on autonomy in the service of profit (Wiener, 1954). In scholarship the limitations on autonomy are inherent in intellectual tradition. As Perkins (1981) notes, scholars work according toestab- lished theories and methods. Creative scholarship is comparatively rare simply because it involves changing well-established theories and methods. The mixture of talent, training, and chance necessary to effect a changein conventional methods and theories is necessarily quite rare (Abelson, 1967; Perkins, 1981). Autonomy, however, is Maracteristic of the arts. In the arts, an authentic statement (a creative product) is possible via the imaginative vision of the artist alone. Aesthetic form is the aspect of art that embodies artistic autonomy, according to Marcuse (1978). Eveninept art is thus, by definition, a divergent view of the world, and authentic art is necessarily creative. This view of creativity helps explain why excellent works of art are much more common than creative scholarship or creativecommercial products. 95 Creativity Contracts 87

PERSONALITY TRAITS ASSOCIATED WITH CREATIVITY

In contrast to the limited research on creative products, analysis of the creative person has been a typical means of investigating cm,tivity. The results of these investigations and speculations are voluminous, if not de- finitive. This research has ranged from Lombroso's (1891) quasi-scientific study of human behavior, through Freud's work in psychoanalysis, to the empiricism of contemporary psychology. This brief discussion can not do justice to the diverse and complex theories developed to account for correlates of creativity. However, it offers an outline of the ideas and research findings that are most relevant to educators of gifted students.

Early Studies of Creative Genius

..t was only when American psychologists began to define intelligence in terms of IQ that creativity came to be treated as distinct from intelligence and even contrasted with it. Until the twentieth century, the concept "creativity" was subsumed by the concept "genius." Lombroso's The Man of Ge- nius and Galton's Heredittny Genius, both published in England in the nineteenth century, made no distinction between the two concepts. Lombroso's characterization of geniuses as physically and emotionally degenerate instigated one of the major controversies in the study of giftedness: the relationship between genius and insanity. Lombroso's list of characteristics (e.g., pallor, stammering, left-handedness, amnesia, and early death) contrasts sharply with contemporary lists, which tend to characterize the gifted in glowing terms (e.g., energetic, outgoing, funny). However, the belief that creativity anti insanity are close relatives has not been dispelled. Psychologists still debate the existence of a relationship between the two (cf. Eysenck, 1983). Noticing that btilliant men seemed to share certain traits, Lombroso proceeded to look for more examples to support his hypothesis. The problem with this method of collecting data is that it could establish only the occurrence of certain traits among men of genius. It couldreveal nothing about the relative frequency of occurrence. It certainly could not establish, as Lombroso claimed, that the traits occur morefrequently in the brilliant than in the normal population. Lombroso's influence must be attributed not to the quality of his argument, but to the concordance of his findings with popular sentiment. Many people were convinced that genius and madness were closely associated; and a spate of biographical case studies, mostly documenting the psychological aberrations of eminent geniuses, were published. Although Francis Calton's English Men of Science and his Hereditan, Genius were among these biographical works, Galton did not subscribe to

6 88 Creativity Cox:buds Lombroso's view. At least he did not see the creative genius as unbalanced. In fact, creative accGrnplishment, in Galton's view, demands emotionalbal- ance. According to Galion, eventhough the eminent poets, artists, and musicians that he studied were ". ..exceedingly irregular in their way of life (Galion, 1869 p. 278), they combined intellect, rigor, and earnestness with "delight in the exercise of their affections." This observation fore- shactowed, and perhaps influenced, contemporary psychologists' findings that creative adults seem to combine psychological vulnerability and exceptional psychological strength.

Psychoanalytic accounts of creativity.Sigmund Freud's concept of creativity as a means or resolving personal, unconscious conflictsformed yet another association between mentaldisturbance and creativity, although Freud did not imply that abnormal needs and conflicts werethe source of creative work. Freudconter-ded that creative endeavor is only one of sc.eral mechanisms for coping aiththe unhappiness and frustra- titm that everyone experiences. The unconscious desirtsunderlyin4 creative activity require sublimation, the redirection of basicpsy-Lhosexual drives so that they are not frustrated by the external weyid (Strac hey, 1961). However, according to Freud, the pleasure of creative workis not accessible to everyone; it requires special dispositions and abilities. C. G. Jung, in reactior to Freud, held that the individual unconscious does not inform the content of creative works, but only of self-indulgent or neurotic exercises. According to Jung, the artist must raise images, not from the personal unconscious, bet from the colleaive unconscious. From this part of the psyche, which he says 1. less accessible than the personal unconscious, come archetypal images that reflect universal experiences rather than personal history. Artists transform these images into symbols that can be understood by their contemporaries (Campbell, 1971). The creative personality is, for Jung. necessarily maladaptive because only dissatisfied misfits are sufficiently inured to conventional ideas and mores to be receptive to the archetypal images of human experience. Jung's account is generally denigrated as mystical, but it nonetheless seems to corroborate the accounts of someartists. Although such accounts do not demand a supernatural source, they are often attributed to one. Consequently, Jung's theory of the source of reative work is regarded by most scientists as an inappropriate line ofinquiry; scientific methods cannot legitimately he applied to mystical accountsof behavior. In Jung's account, artists must be introverted in order to resist external influences. This resistance to external influences and attention to internal ones is essential to the creative process in Jung's view. Empirical investigation does support the hypothesis that artists are introverted (e.g., Barron, 1963; Barron, 1969), but these findings do not, of course. validate Jung's theory of the source of creativity. 9 7 Creativity Constructs 89

Humanistic psychologists' view of the creative person.Alfred Adler belongs to the psychoanalytic tradition, but his view of personality, more optimistic than the traditional Freudian view, is a precursor ofhumanist psychology. Adler's psychology posits a kind of psychological upward mobility, that is, a drive toward supet iority or, in humanistic terms.self-actual- ization. Adler believes that creativity is an individual's healthy response to physical inferiorities. According to Adler's theory, creativity is a formof compensation for inferior physical organs: Poets and artists are compensat- ing for inferior visual structures; musicians for auditoryweaknesses; and actors, singers, and speakers forinferiority of the vocal apparatus (Ans- bacher & Ansbacher, 1956). This theory of creativity, thoughcertainly interesting, is, like many psychoanalytic theories, very difficult to test empirically. Much of Adler's influence is in: irect and is seen in thework of Abra- ham Maslow (1954), who elaboratel the concept of theself-actualizing individual. Maslow's deveiopment oi this idea included a hierarchyof needs basic to all humans (i.e., a need or satisfaction of primaryphysical drives, a need for physical and psychological safety, and a needfbr a sense of belonging and love). In addition to these fundamental needs,Maslow hypothesized the human need to actualize potential. Using a casestudy approach imilar to that used by Lombroso and Galion, heidentified characteristics of persons whom he considered to be self-actualizing (e.g.,Al- bert Einstein, William _lames, Eleanor Roosevelt). Thecharacteristics he found these people to share included (1) a more efficient perceptionof reality and more comfortable relations with it; (2) acceptanceof self, others, and nature; (3) spontaneity;(4) it problem-centered rather than ego- centered otientatitm; (5) preference for solitude and privacy: (6) autonomy; and (7) creativeness(Maslow, 1954). The creativeness to which Maslow refers when he describesself-actu- alization is different from the "special-talent creativeness of the Mozart type," (Maslow, 1954, p. 223). As he points out inhis description of self- actualizing individuals, his subjects did not show special creativitythrough writing fiction, painting, or composing music; rather, theymanifested an originality of outlook and an open, spontaneous approach toboth their work and their play. According to Maslow, this style, born outof satisfaction of basic physical and psychological needs, lends freshnessand inventiveness to their ideas. Maslow's findings provide a rationale for student-centered teaching. counseling, and a psychologically supportive classroom environment: strategies that are conducive to openness and inventiveness. There is no evidence from his work, however, that these strategies are effective in developing the creative talent needed to produce outstanding novels,musical compositions, poems, paintings, and dramas. Application of Maslow's find-

9 es 90 Creativity Constructs ings to the education of gifted students often disregards the distinction between a flexible, spontaneous (i.e., "creative") approach to life and the actual creation of significant scholarship or works of art. However, Maslow himself was not optimistic about influencing the making of excellent products:

We may as well face the fact that the so-called geniuses display ability that we do not understand...they seem to be specially endowed with a drive arid a capacity that may have rather little relationship to the rest of the personality and with which, from all evidence, the individuals seem to be born. (Maslow, 1954, p. 233)

Whether giftedness is innate or learned is still a matter of controversy, but various sorts of research do suarst a relationship between some personality qualities and creative genius.

Recent Research Findings on Creative Adults Humanistic psychology cultivated an interest in the oxceptionally healthy personality and in people who were unusually competent. During the 1950s and 1960s, several studies set out to identify cogniti%e and affective traits of highly reputed artists, scientists, architects, mathematicians, and businessmen. These studies have much in common with the approaches of Maslow, Galton, and Lombroso to the study of creativity. How- ever, the recent studies differ from the earlier biographical studies in several ways: Subjects were selected on a more random basis, sample size was larger, and measures of personality were more objective. Among the variables found to distinguish creative persons from others were their interest in abstract ideas, lack of interest in socializing, unhappy childhood experiences, emotional intensity, emotional strength, and emulation of nonconformist behavior.

Interest in abstract ideas.Creative artists, writers, mathematicians, scientists, and architects prize aesthetic and theoretical pursuits (Perkins, 1981). On the Allport-Vernon-Lindzey Study of Values (1951)a measure comparing relative interest in theoretical, economic, aesthetic, social-political, and religious ideascreative persons usually score highest on the aesthetic and theoretical scales (MacKinnon, 1978). Architects tend to score highest on the aesthetic scale and second highest on the theoretical scale: scientists score highest on the theoretical scale, but their aesthetic score is nearly as high; and mathematicians score about equally high on both scales (MacKinnon, 1975). Gifted art students' aesth :tic scores are extremely high and their theoretical scores next highest (Getzels & Csikszentmihalyi, 1968). Commitment to aesthetic and theoretical issues was also shown to be

or Crrativity Casstructs 91 one of the most salient characteristics of a groupof renowned writers (e.g., Truman Capote, Norman Mailer, MacKinlay Kantor, Frank O'Connor, and Kenneth Rexroth) who participated in a study on creativity. Barron (1963, pp. k4"-243) was impressed with the intensity of the authors' ". .. profound commi:.vient to larger meanings of an esthetic and philosophical sort. .." It really L...;ald come as no surprise, however, that women and men who spend most 1:f their livesworking at art, music, mathematics, or literature report that they are devoted to the essential content of their disciplines.

Lark of interest in socializing.Those creative persons who score high on interest in aesthetics and theoretical issues usually scorelow on the social scale of the Study of Values. MacKinnon (1975) found that about two- thirds of a sample of creative writers, architects, and scientists were substantially more introverted than their less creative peers. Scientists who are rated high in creative ability often describe themselves as loners (Andrews, 1975; Roe, 1952). Artists, too, are more introverted than the rest of the population (Gotz & Gotz, 1973; 1979). There is little question that introversion predominates among the highly creative. What is not clear is whether their introversion results from insecurity about sucial interactions, from hostility toward others, or from enjoyment of solitude and intellectual pursuits. Some studies stress that scientists are socially competent, but simply prefer intellectual to social challenge (e.g., Roe, 1952); other studies suggest that insecurity contributes to scientists' introversion (Andrews, 1975). Thetendency toward introversion may not result solely from either factor; it may, instead, reflect both preference for solitude and alienation. Introversion may arise from and support the process of creation (Getzels. 1979: hut cf.Osborn, 1953).

Unhappy childhood experiences.Highly creative adults are more likely to report unhappy childhood experiences than are less creative adults (MacKinnon, 1978). Whether or not these self-reports are accurate is open to question. The self-reports of highly creative adults are not sodefensive as those of less creative persons(Barron, 1969), and this fact alone may account for their reports of unhappy childhoods. MacKinnon(1978) maintains that creative subjects' reports of unhappy childhood events corfirasted with their apparently favorable childhood conditions. On the other hand, Gardner (1973) and Simonton (1984) report that a surprising percentage of majorresearch scientists lost a parent before the age of ten. Roe (1943) found that six ofthe 20 professional artists she studied had experienced the death of a parent or sibling during their teens, six of the 20 were socially isolated, and three had serious illnesses when they were children. However, this information is of little significance unless we find that such crises occur with greaterfrequency or have a greater

1 0 0 92 Crrativity Constructs effect than in the normal population. As fir we know, noneof the studies on creativity has compilednormative data with which to compare the experiences of creative subjects.

Emotional intensity.Most creative artists and writers report intense emotional responses, awareness of their responses, ard acceptanceof their emotions. The highly creative authors studied by Barron(1969) expressed more extremes of emotion thandid other authors who served as a control group. The creative authorsalso reported a more vivid dream life and more nightmares than the control group.They tended to he more opwn to experience and less judgmental; and they attached moreemotional significance to objects and events.This intensity of emotional response, sensitivity to meaning, and candidness abouttheir feelings may help account for artists' reputation for insanity and their high scores on measuresof emotional disturbance. Highly creative groups showed more psychopathology onthe Min- nesota Multiphasic PersonalityInventory (Hathaway & Meehl, 1951) than did other members of their profession (Barron, 1969).This effect was particularly noticeable among creative authors, whoscored high on all MMPI measures of psychopathology (e.g., subscales designed toidentify schizophrenic, depressive, hysterical, and hypochondriactendencies). Pro- fessional artists earn much higher psychoticism scores on the EysenckPer- sonabty Questionnaire (Eysenck & Eysenck, 1976) than do subjectswho are not artists, and male artists scorehigher on neuroticism as well (Gotz & Gotz, 1979). Gifted art students score higher on neuroticismthan other students (Cotz & Gotz, 1973). A number of studies have found a positive correlationbetween creativity and schizophrenia in different members of' the samefamily. In a brief' review of the literature, Eysenck (1983) reports thatabout half of' the children (raised by foster parents) of' schizophrenic mothersexhibited psy- chosocial problems; the other half', who seemed emotionallystable, had uncommon art talent. AlthoughEysenck and the researchers whose work he reviews tend to attribute creativity and schizophrenia to a common genetic component, that conclusion is not a necessaryimplication of their findings. The literature on creativity and insanity does not document serious emotional difficulty in creative persons. One explanation fOrthis finding is that researchers document the characteristics ofsurcemlidcreative persons. Another possibility is that the emotional intensity. candor, andmmconfor- mity of creative individuals are exceptimal. Theirvalues differ from the ordinary, and they do not hesitate to admit it. Since labels ofemotional disturbance reflect traditional values, individuals whoreject those traditions are more likely than average subjects to be categorized aspsychotic or neurotic regardless of their psychological competence. Crrativity Constructs 93

Emotional health.Barron (1963) came to the conclusion that highly creative individuals are both sicker and healthier psychologicallythan other people. That is, thoughcreative persons appear more troubled psychologically, they also appear to have greater psychological resourceswith which to deal with their problems. In contrast to the typical patternfor subjects who score high on the Minnesota MultiphasicPersonality Inven- tory psychopathology scales, creative subjectsalso score high on the MMPI ego-strength scale, a measure that usually correlatesnegatively with psychopathology scales (Barron, 1969). The creative subjectsalso score high on the California PersonalityIndex, a result that suggests they are flexible and persistent. Many informal observations of creative personsdocument their self-discipline, commitment, and independence (e.g.,Gahon, 1869/ 1962; Perkins, 1981).

Emulation of nonconformist behavior.Our review of research confirms the stereotype of the sensitive, rebellious artist(cf. Perkins, 1981). However, the stereotype is probably not just the resultof artists' behavior, but also a cause of it. As Barron (1969) notes in discussingthe MMP1 scores of creative writers, being labeled "eccentric," or even"mad as a hatter" is considered a comphmetu when it is used todescribe someone who is creative. Children who aspire to artistic or scientific achievement arelikely to be aware of the stereotypes of the highly emotionalartist and the coldly intellectual scientist. They model their behavior accordingly.This phenomenon is suggested by art students' commentsduring interviews published in Barron's Artist.% in the Making. When asked if theythought of themselves as artists, two students replied as follows:

Id rather sav I'm craiv than say I'm an artist. Because"crazy" means untold vision, infinite vision. Whereas "artist" means von have afew visions maybe. (Barron. Kraus, & Cmti. 1972, p. 25.) think the ultimate artist is a magician, but there arelots of stages in between. I live a Bohemian lik. (Barron et al., 1972, p. 20)

When asked if. as a child, she felt difkrent fromother children, one of the students replied, "I'd like to think that Idid" (Barron et aL, 1972, p. 29). When asked who had affected hisinterest in art, another student responded, "My cousM. fie's teaching painting now. He's a topartist. Ile's insane, but everrme appreciates his work"(Barron et al., 1972, p. 22). Cr :_ative Mdividuals value being different,and they actively deny conventional behavior and values (e.g., Marcuse, 1978). AsMaddi (1975, p. 183) comments, many creative children andadults find opposition to their ideas stimulating and enjoy the role of a "lonely searcherand crusader for fUlfillment amidst a multitude of drones."

1 2 94 Creativity Constructs

Personality Traits of Children Who Are Divergent Thinkers Children who score high on divergent thinking tests are, by defini- tkm, capable of unusual and clever responses to questions or assigned tasks. They are generally nonconformists, risk takersintellectually playful, humorous, and known for their wild ideas (Torrance, 1962). Among bright children. Getzels and Jackson (1962) found that highly divergent thinkers value a sense of humor more than do convergent thinkers, that they tend to be less emotionally close to their parents, and that they seem less concerned about money. Divergent thinkers in the classroom.Teachers tend to prefer bright children who score highest on tests of convergent thinking to those who score highest on tests of divergent thinking(Getzels & Jackson. 19(32). In fact, highly divergent children are often in conflict with both teachers and peers. Torrance (1963) reported that highly divergentchildren with whom he worked were given little credit for their contributions to group problem- solving sessions. Onlythose who persisted in the faceoftheir peers' hostility finally won recognition by the group. Divergent thinkers generally prefer to learn in settings less highly structured than those preferred by other children (Torrance. 197.). If we think of divergent and convergent thinking as learning styles rather than kinds of thinking, it is easy to see that the convergent style is more compatible than the divergent style with most classmom instruction.

interactionsbetween IQ and divergent thinking scores. Wallach and Kogan (1965) categorized fifth-grade students as belonging to one of four groups: high-IQ, high-divergent thinking; high-IQ,low-divergent thinking; low-IQ. high-divergent thinking; and low-IQ, low-divergent thinking. In comparing the classroom behavior ofchildren in the four groups, the researchers found differences by group and by sex. Girls who scored high in IQ and divergent thinking were the most self-confident group of girls, They showed the least tendency to deprecate themselves or their work. They were more actively sought out by their peers than were girls in any other group,and they also sought out compan- ions more actively than did other girls. They showed the highest levels attention, concentration, and interest in academic work. They were also high in disruptive, attention-getting behavior; but this seemed to be a function of their enthusiasm for learning. 'The high-IQ girls with low-divergent thinking scores were also self- confident, had a long attention span, had the ability to concentrate, and were sought after by others. They did nottend to seek companionship. however. They were the least likely of any group to seek attention in disruptive ways. Creativity Coattruets 95

The low-IQ. low-divergent thinkinggirls seemed to compensate fbr their poor academic performance byactivity in the social realm. The girls who had low IQ scores andhigh divergent thinking scores seemed to he at the greatest disadvantagein the classroom. They were the least confident and the least sought after group.They avoided companion . ship and tended to deprecate themselvesand their own work more than any of the other groups.They frequently engaged in disruptive,attention- getting behavior, which the experimentersregarded as a form of protest. Personality differences among the boys were not sodistinct as among the girls. Nonetheless, boys whohad high IQ scores and low divergent thinking scores showed the least anxiety,whereas anxiety was highest for the low-IQ, low-divergent thinkingboys. Probably because of sex-role stereotypes that allow morebizarre behavior from boys, low-IQ. high-divergent thinking boys seemed tohave less difficulty in the classroom thantheir female counterparts.

Children who resembk creative adults.The characteristics of creative adults often guide the identification ofgifted children. Several checklists, for example, have been compiled forteachers' use in identifying children with unusual creative ability. The GroupInventory for Finding Talent, or (Rimm & Davis. 1980), and the Group InventoryFor Finding Inter- ests, GIFFI (Davis &Rimm, 1982), are composed of items representing interests, attitudes, and family traitscorrelated, either negatively or positively, with creativity. The Renzulli-liartmanScales for Rating Behavioral Characteristics of Superior Students (Renzulli &Harm ia 1971) include a section composed of items describing traitsassociatedith creativity. Although these and similar inventories areuselul for research purposes and for finding outabout children's interests and background,their use as predictors ofoutstanding creative talent is based onquestionable assumptions. Both identification instrumentsand instructional programs predicated on this identification model Psstime acausal relationship between personality traits andcreative production. The existence of such a causal relationship is only speculative, andthe strength of the hypothesized relationship is entirely unknown.

CREATIVITY AND INTELLIGENCE

earlier in the chapter. we saw that interest ineminence prompted the study of both intelligence and creativity. Wealso saw how the study of creativity was derived fromthe study of intelligence. In this part of thechapter we will examine the specific ways in whichthe measurement of creativity is related to the meavurement of intelligence. Inexamining this relationsL;we will explore the notion of "creativepotential" and compare it to the notion

104 96 Creativity Cvastructs of "intellectual potential." We will consider the degree to which measures of creativity can predict eminence and the degree to which they can predict school achievement.

The Creativity Construct A measurement construct operationalizes a concept statistically. A construct is meaningful if it (I) samples the components of an attribute and (2) distinguishes between the attribute and other attributes. The extent to which a particular test conforms to a theory explaining an attribute is a measure of the test's construct validity. One way to determine the construct validity of a test is to measure the degree to which the construct that it samples is distinct from other constructs. Once a test has been found to measure a discrete construct, other tests that purport to measure that construct will be correlated with the original test to determine their concurrent criterion-related validity. (See Kubiszyn & Borich, 1984, for a succinct discussion of validity.) The construct validity of creativity instruments depends on the distinction brtween "creativity" and other mental attributes. However, itis difficult to interpret validity studies because the construct, -creativity," has been formulated in at least the following four ways: (1) as a noncognitive trait (e,g,. Welsh, 1975), (2) as a combination of fluency. fleAibilits Aiwa- tion, and originality subskills (e.g.. Torrance. 1966), (3) as tlw a..)ility to associate remote elements (Mednick. 1962). and (4) as a combination of fluency and uniqueness subskilh (Wallach & Kogan. 19(15), After looking at these four "creativity" constructs, we will examine the results of studies to distnguish each of' these constructs fi-om the intelligence construct.

Creativity as a non-cognitive trait.In one view reativity is a noncognitive (i.e.. affective) trait. Such a trait falls outside the domain of the intellect and, in theory, has little to do with intellectual functioning. An aflective trait explaining creative performance ((mid relate to any of various aspects of personality, such as motivation (see, e.g.. McClelland, Atkin- son, (;lark, & Lowell. 1953; Pearlman, 1983), or it might relate to a discrete noncognitive attribute. Welsh (1975) terms one such attribute "origence.- In the preceding section of this chapter. we summarized findings about the personality traits of creative individuals. The discussion noted the use and limitations of personality checklists to evaluate children's creative potential. Such checklists are not psychometric instruments and do not represent measurement constructs that can be contrasted with the IQ construct. Various psychometric instruments have, however, been used to nwa- sure creativity. Most prominent among these have been projective tests that are sensitive to the noncognitive aspects of creativity. A test of this type that

1 15 Crrotivity Constructs 97 has been used extensively in research is the Welsh Figure PreferenceTest (1949). This test has been revised and is now called the Barron WelshArt Scale (1963). It consists of 60 cards, each showing a simple linedrawing. Subjects are asked to view each drawing and respond by saying either"like" or "don't like." The level of asubject's "origence" is based on his or her preference for particular sorts of figures. High origence is associated with preference for more complex and asymmetric figures (Welsh, 1975).

Creatsvity a.sfluency,flexibili4y, originality, and elaboration.Guilford and Torrance view creativity as a constellation of cognitive traits,including fluency, flexibility, originality, and elal)oration. These skills comprisedi- vergent thinking, the construct that Guilford and Torrancebelieve reflects creative potential. This view of creativity is operationalized in the Torrance Tests of Creative Thinking (1966). Fach of the verbal and figuralsubtests of the TTCT is scored fbr fluency, flexibility, and originality;and several subtests are also scored for elaboration (see Box 4-2). This formatdiffers considerably fibm the fbrmat of many tests that purport to measure difier- ent subskills (e.g.. Wechsler, 1974). Mostoften, tests use separate tasks to measure different skills or abilities.Because it measures several skills or abilities with the same task. Torrance's procedure obscures thepossible differences between these skills. According to some researchers,students' fluency scores are likely to confound both their originality and theirflex- ibility scores.

BOX 4-2Creativity Subskills in tbe Style of Torrance

Among the most frequently used tests of creativity are the Torrance Tests of Creative Thinking (originally published as the Minnesota Tests of Creative Thinking). The original Torrance battery included verbal and figural tasks that were scored for fluency, flexibility, and originality. According to Torrance (1963, p. 95), fluency is "the ability to produce a variety of ideas or hypotheses concerning possible solutions to problems." Torrance (1963, p, 96) sees flexibility as "the ability to adapt to changing instructions

.. to use a variety of approaches." He views originality as theability to produce uncommon responses; remote, unusual, or unconventionalassociations" (Torrance, 1963, p. 96). In the 1966 version of the battery, Torrance included an additionalfigural subskill of creativity, This skill, elaboration, has been defined as "the ability to expand,develop, particularize, and embellish one's ideas, stories, and illustrations" (Callahan & Renzulli, 1981, p. 395). Below are some examples of the types of items used to assess studc,ts' verbal fluency, flexibility, originality, and elaboration.

Unusual Uses. list all the possible ways in wh.c h you might use a safety pin. Unusual Questums. list os many questions s you can about safety pins. Try to think about safety pins in ways that people usually do not consider them. 98 Creativity Cenutructs

Product Improvement: Here is a toy fire engine. Describe all of the ways in which it might be changed to make it more fun to play with. Ask and (;uess. Look at a line drawing showing two people engaged in what looks like a robbery of a bank; then write all of the questions evoked by the picture. Next, list as many possible causes of the events in the picture as you can imagine. Finally, list as many possible consequences of the events in the picture as you can imagine. just Suppose: lust suppose all of the fresh water on earth became salty. What would hapren? How would this change life on earth?

Creativity as remote association.Mednkk proposed a creativity construct based on the idea that c reative individuals are those who are best able to join remote elements into "new combinations which either meet specified requirements or are in some ways useful" (1962, p. 221). According to Mednick, creative individuals are those who can join remote elements. Therefore, he devised a test that requires students to find the one word that links progressively more remote associates. A sample item from the Remote Associates Test (Mednick & Mednick, 1967) asks students to supply the one word that links the f011owing three words:sore, shoulder,and Awral. The answer is "cold." As this sample item indicates, the items on the RAT reflect a theory of creativity that views it as both cognitivr and con- yrrgenl.

Creativity as fluency and uniqueness.Wallach and Kogan (1965) developed a creativity test based on a construct that intended by delinthon to distinguish creativity from intelligence. This construct "limitIsj the term rreath,ity to a narrowly defined set of variables, thereby identifying creativity with a single factor" (Anastasi & Schaefer, 1971, p. 115). 1.ike theor- rance Tests of Creative Thinking, Wallach and Kogan test battery includes verbal and figural subtests. The scoring procedures and attendant difficulties are similar to those discussed above in reference to the TiC. The Wallach and Kogan battery. however, differs from the TTCT in two important ways. First, the Wallach and Rogan battery is administered without time limits. Second, "uniqueness" scores on tlw Wallach and Kogan test are based on local norms rather than on norms derived from the testing of a representative sample of students. A response is judged to be "unique" only if no other student in the Aamr frAt-administration group has given the same response. This procedue makes comparison of different groups. and thus development of reliability and validity coefficients, virtually impossible. The Relationship between Creativity and Intelligence Determining the correlation between creativity and intelligence is necessary in order to distinguish between these constructs. Theorists who maintain that creativity is a construct discrete from general intelligence. Creativity Contracts 99

(e.g, Torrance, 1984; Wallach & Kogan, 1965) supporttheir view by showing the low correlations between creativity ;Indintelligence. Psychologists who maintain that creative potential is stronglyinfluenced by intellectual potential (e.g.. Griffith & Clark, 1981; Simon & Ward,1973) support their view by showing the high positive correlation betweencreativity and intelligence. Still other theoristswho view intelligence as a necessary but not a sufficient conditim for creativity (e.g., Getzels & Jackson,1962; Schubert, 1973) support their view by showing the weakeningof the correlation above a certain point on the IQ scale (cf. McNemar,1964).

The low correktion between creativity and intelligence.According to Fox (1981, p.23I), "researchers have amplydemonstrated that IQ and achievement tests measure abilities, skills, andpersonality traits that are essentially dif ferent from those associated with creativity."Fox (1981) bases this conclusion on a review of numerous studiesin which the correlations between creativity and IQ were found to be verylow. Several of the studies discussed by Fox (e.g.,McKinney & Forman, 1977; Metcalfe. 1978; Ward, 1975) used the Wallachand Kogan test as the measure of' creativity.This test was &signed to be factorially distinct from measures of intelligence.Although low correlations between this test and measures of intelligence ctmfirm the successof' Wallach and Kogan's test design, these resuits maybe insuffic;ent to validate a measurement construct that distinguishes"creativity" from "intelligence" (Anastasi & Schaefer, 1971). Because of the limited sampleof behaviors measured on the Wallach and Kogan test, this instrument should notbe construed as a measure of "vreativity." A morereasonable assertion is that the Wallach and Kogan battery primarily measures ideationalfluency, a skill that has bven shown to be distinct from intelligence (see, e.g.,Hocevar, 1979). By definition. "creativity" based on anoncognitive wnstruct differs considerably from any cognitive construct, includingthe rognitiw constructs of "creativity," "intelligence," and "readingcomprehension." Welsh (1975). for instance, reports a 0.06 correlafimbetween the Revised Art Scale (Welsh & Barron, 1963) and the Concept MasteryTest (Ferman, 1973). This low negative correlation suggests that variancein performance on a test of vocabulary knowledge has verylittle to do with preference for line drawings of a particular type. Viewed in these terms, we can seethe limited practical value of the statistic cited by Welsh (1975). Inthis case, neither the measure of intelligence nor the measureof' creativity is sufficiently complex to enhance our understanding of either construct.

The high correlation between creativity and intelligence.Several psychologists suggest that there is a close relatiomhip betweenintelligence and creativity. According to McNemar (1964), for example,methodological problems in many studies correlating creativity andintelligence often obscure the interrelatednessof these constructs. McNemar (1964, p. 879) 100 Creativity Construe& claims that "the corre/ations !between IQ and creativity tests] are generally far higher than those found in typical studies with range restrictions." The Remote Associates Test (Mednick & Mednick, 1967) is the creativity test most likely to be highly correlated with intelligence. This test, which is both cognitive and convergent, has been compared with IQ tests in a number of studies. Welsh (1975) reports a rangeof.rrelations between the kAT and various intelligence tests of from 0.19 to 0.55. The TTCT (Torrance, 1966), however, claims to measure di.... nt thinking and should therefore be less highly correlated than the ILY 4h measures of intelligence. Nevertheless, Wallach and Kogancite research to show that the correlations between IQ tests and the Torrance scales are as large as intercorrelations either among TTCT subtests or among IQ subtests. Thus, rrur subtests and IQ subtests are similar enough to represent aspects of the same psychological construct. Other researchers have noted similar results in reference to the Wal- lach and Kogan battery. Based on a study using the Wallach and Kogan test, Griffith and Clark (198), p. 233) conclude that "even amongsubjects of relatively low creative ability, intelligence has an important impart on sonie aspects of creative responding."

Measurement cautions,Interpretation of the correlation studies must be tempered by an understanding of the limitations of' the tests used to represent the two constructs. "Intelligence" is operationalized as IQ on any one of several tests that vary as to the number and kind ofbehaviors sampled. "Creativity," on the other hand, is operationalized on tests that vary not only as to the specific sorts of behavior sampled but also as to the domain from which these behaviors are sampled. As we have seen above, theorists do not agree about the mture of the creativity construct. Further, the creativity tests developed to date are neither as reliable nor as valid as most IQ tests (see, e.g., Thorndike, 1972). Simon and Ward (1973) suggest a reasonable interpretation of the apparentlY emitradictory data obtained in the correlation studies. hey ttmclude that **significant positive correlations exist between certain kinds of creativity measures and certain kinds of intelligence tests, and .. !that] insignificant positive correlations exist between other creativity measures and other kinds of intelligence tests" (Si.non & Ward, 1973, p. 72). Sitaon and Ward's conclusion is supported by Hocevar (1979, 1980) who identifies "ideational fluency" (i.e., the ability to generate many ideas) as the "creativity" factor that is distinct from intelligence. In astudy using three of Guilford's creativity subtests. Hocevar (1979. p. 194) finds that Ideational fluency is the factor that causes creativity tests to he highly intercorrelated and factorially distinct from intelligence tests ...but this does not mean that the broader category of creative thinking will be fat-- torically distinct when the effect of ideational fluency is controlled."' Thus, Creativity Constructs 101 creativity tests that are highlycorrelated with IQ tests (e.g., RemoteAssoci- ates Test, Mednick &Mednick, 1967) probably do notsample ideational fluency to a large degree. Thecreativity tests that are less highlycorrelated with IQ tests (e.g., Torrance, 1966;Watlach & Kogan, 1965) probablyrely heavily on ideational fluency to representcreativity.

Intelligence as necessary but not sufficientfar creativity.The notion that a certain level of intelligenceis a prerequisite for creativityis supported by the common sense viewof the creative person (e.g.,Perkins, 1981). This view also recognizes thefact that not all intelligentpeople produce creative work. Althoughthese notions about the natureof the creative person seem to make senseintuitively (see, e.g., Dacey & Madaus, 1971), they are not supported conclusivelyby the empirical evidence. Interest in a threshold hypothesisconcerning the relationship between intelligence and creativity was generated as aresult of Getzels and Jackson's (1962) classic study. This study seemed toshow that creativity and IQ were quite distinct at the upper ranges ofintelligence. McNemar (1964, p. 879) stated the hypothesis more expliciilywhen he wrote, "at the high IQ levels there will be a very wide rangeof creativity, whereas as we godown to average IQ, and ondown to lower levels, the scatterfor creativity will be less and less." Although several studies (e.g.. Bowers,1969; Dacey & Madaus, 1971; Schubert, 1973) seem to confirm thishypothesis, other studies (e.g., Guilford & Christensen, 1973;Marjoribanks, 1976) dispute the threshokl hypothesis. According to Daceyand Madaus (1971, pp. 215-216) researchers arrive at varyingconclusions about the threshold hypothesisbe- cause of the dubiousvalidity of the currently popular measur-sof divergent thinking. Untilthe technical problems associatedwith tests of creativity are resolved, it is unlikely thatresearchers will be able to quantify the relationship between IQ andcreative potential.

Creative Potential and Intellectual Potential The definitions of the two termsintellectual potential and creative' potential may help clarify an essentialdifference between measures of creativity and measures of intelligence. InChapter 3 we saw Ifni: the let in intelkctual potential linked the general notion ofintelligence with the narrower measurement construct, IQ.Because IQ tests effectively predictschool achievement, they can beused to locate children likely to performwell or povrly on academic tasks.The intellectual potential of school children,therefore, refers to their potential for academicachievement. Several theorists (e.g., Getzels & Jackson,1962; Torrance, 1963) have described creativity tests as measuresof potential. They maintain that although creative potential differs fromintellectual potential, it nevertheless is

1 0 102 Crrativity Contracts as effective as IQ in predicting academic achievement, They suggest that above a certain IQ level, performance on measures of creativity predicts academic achievement as well as or better than performance on IQ tests. According to these theorists, creative potential, like intellectual potential, refers to students' potential for academic achievement. Getzels andckson (1962), in particular, hypothesized that high creative potential explained the apparent "overachievement" of some students.

Do creativity tests predict academic achievement?In order to support or refute the claim that creativity tests predict academic adlievement, several researchers (e.g., Bowers, 1969; Bruininks & Feldman, 1970; Mar- joribanks, 1976) have conducted studies of the relationship betweenper- formance on creativity tests and performance on tests of academic achievement. Rather than clarifying this relationship, however, these studies have confused the issue. For example, Marjoribanks (1976,p. 117) found that "for certain academic subjects rreativity is related to achievementup to a threshold level of intelligence, but after the threshold has been reached creativity is not associated with further increments in achievement." This finding is the inverse of the relationship suggested by Getzels and Jackson's (1962) data and hypothesized by McNemar (1964). To date no consensus has emerged about the usefulness of creativitv tests in predicting academic achievement. If creativity tests are found to correlate highly with IQ tests, then they are likely to correlate with academic achievement. If creativity tests are found to correlate minimally with IQ tests, then it is unlikely that they will correlate with academic achievement. In either case, educators would be unwise to use creativitytests in place of IQ tests to predict academic achievement since creativity measuresseem to lack both reliability and validity (Thorndike, 1972; but cf. Torrance, 1984).

Do creativity tests predict creative achievement in adulthood?Ac- cerding to several writers (e.g.. Rekdal, 1979; Torrance, 1980, 1984),measures of' creative potential predict creative achievement or eminence in adulthood. This claim is contested by Goodwin and Driscoll (1980) who report that, in general, creativity tests have very poor reliability and validity. The claim that a test measures potential dependson its predictive validity, and creativity tests that have neither adequate construct validitynor demonstrable predictive validity will not help educators identity the children most likely to be creative adults.

SCHOOL RESPONSES TO CREATIVITY

Among teachers interest in creativity seems to have been prompted historically by the observation that some pupils showed academic promise despite their troublesome antics (Torrance, 1984). Because of their behavior, how- 1 1 1 Creativity candela-fa 103 ever, it was sometimes difficultfor these students to realize their promise. Thoughtful teachers in the early part of the century seem tohave been asking, "What phenomenon can account for this constellationof apparently contradictory traits?" The redefinition of the term "creativity"after 1950 seer led to hold promise as apossible description of such students' behavior. Researchers (e.g., Torrance & Myers, 1971) tried notonly to identify highly creative students who were at risk in schools, but totrain teachers to respond to such students more flexibly.

Spiritual Schooling As we implied above, creativity training is considered by some(e.g., Adler and Maslow) to provide a direct route to a moreholistic, more healthy, more self-actualizing approach to life. Others haveexpanded this sense of creativity to include someof the data on brain hemisphericity (Myers, 1982), some of the constructs of psychotherapy(Gowan, 1978; Huff, 1978), some of the techniques of Eastern mysticism(Gowan, 1978; Rose, 1979), and an appreciation of the altered perceptionsof psychedelic experience (Huxley, 1962). -1-he leading spokesperson forthe most successful creativity training program has noted that his programtries "to study and relateeven to 'force' relationshipswith all areas andapproaches" (Parnes, 1975, p. 24). Such a synthesis is a system of belief. It isgrounded on faith, and it represents a largely spiritual view of life. Torrance wanted earnestly to unlock the creative potential of all children (Perkins, 1981; cf. Torrance, 1963, 1984; Torrance & Myers,1971). Other proponents of creativity training (e.g., Sisk & Bierly, 1979) came increasingly to believe that the special methods that benefited highlydiver- gent students would benefit all students. The expectation that schools could liberate the spirits of allstudents was a sentiment shared by manyother educar ,.s (e.g., Lyon, 1971) in the 1960s and early 1970s. Part of the support for such a view no doubt came from articulate students who dissented during the Vietnam War (Tannenbaum, 1981). Creativity training, with its emphasis on divergent thinkingand its promise of toleration for the behavior and lifestyle of nonconfo-mists seems to have found a niche in the progressiveideology of that era. Human beings do need a sense of their wholeness. It is questionable, however, whether very many of us can agree on just what sort of spiritual program would be appropriate for allchildren. Creativity training, however, may appeal to some educators as aspiritual program that is appropriate for all children. Certainly most teachers would like their students to be happy, healthy, and productive. But construction of a sense of wholeness, of one's place in the universe, is a very private matter. Moreover, a gooddeal of research also suggests that the kinds of reforms imagined by Torrance and others probably depend on social changes that educators are powerless to 112 104 Creativity Constructs make (e.g., Jencks et aL, 1972; Wilcox, 1982; Meyer, 1977). Thedegree to which success in such reform efforts is possible is limited by requirements for conformity and compliance in schools and by the financialconditions that govern the way schools are run.

Creativity Training in the Schools Schools are run as businesses are run, to provide a functional service as cheaply as seems reasonable. Part baby-sitting agency(a valuable role), part socializing institution, they mirror the structure and needsof society at large (Bowles & Gintis, 76; Jencks et aL, 1972; Wilcox, 1982; Meyer, 1977; Sizer, 1984). Many good things can happen in the classroom, regardless of the determining influence of social need, yet there are reasonable limits to what ought to be planned. The preceding discussion indicates that concern for bright, troublesome children is warranted,but that too grand an agenda will not serve such childrenor anyoneverywell. It is possible to locate a group of' children who perform very well On tests of divergent thinking. There is some question, however,whether (1) the tests reliably identify the sort of troublesome student described by Torrance (1984), or (2) the tests prompt educators to provide such students the support they need to realize their academic potential. We do know that poor divergent thinkers can improve their scores ondivergent thinking tests rather quickly (Torrance, 1984). Such an improvement does not turn formerly poor divergent thinkers into troublesome students, of course, nor would we want it to. This backwards logic, though, does indicate the degree to which it is possible to forget the beginnings of interest in creativityin the schools, and to confound improved divergent thinking with improved academic performance (cf. Getzels & Jackson, 1962; Torrance, 1984). It should be apparent that, if the definition and measurenwnt of creativity is problematic, so too is creativity training. This observation is not meant to suggest that it is impossible to teach the skills of divergent thinking. Indeed, research indicates that the subskills of divergent thinking can be taught (Rose & Lin, 1984; Torrance, 1984). As Perkins (1981) notes, however, improvement in children's scores on tests of divergent thinking hardly substantiatesthe cultivation of rreativ. y. Unresolved questions pertain to the place of divergence inclassrooms, to the role of improved divergent thinking in cultivating significant contributions in the arts and sciences, to the relationship of conformity and creativity in general, and o the comparative meaningfulness of the term "creativity" itself (Perkiii.,, 1981). For a review and critique of classroom use of creativity training programs, including the Parnes-Osborn method of creative problem solving (e.g., Osborn, 1953; Fames, 1981) and Gordon's method of synecfics (Gordon, 1961), see Howley et al., 1986, pp. 171-181.

1 1 3 Creativity Constructs 105

SUMMARY

This chapter reviewed the concept of creativity from various perspectives. It explored definitions of creativity based on the notion of divergence and examined the role of novelty in definitions of creativity. The discussion emphasized the importance of evaluating the nature and quality of the creative product rather than the cognitive or affective characteristics of its creator. The chapter summarized findings about the characteristics of creative individuals. However, it also cautioned against applying generalizations that result from this research. The intelligence of creative individuals was examined in some detail in the discussion of the creativity construct. This discussion considered the assessment problems resulting from the use of tests that purport to measure creative potential. Finally, the chapter included a brief consideration of the role of creativity training in public schools. Both the intent and the actual experience of creativity training were evaluated in light of unresolved questions about the natureof creativity, its relationship to conformity, and its proper nurture.

1 1 4 Adulthood: Eminence, Leadership, and Careers

I.Focusing Questions II.Concern for the Adulthood of Gifted Students III.Definitions A.Leadership 1.Acceptable and unacceptable models 2. Identification B. Eminence 1.A continuum of talent C.Studying the Relationships among Rare Phenomena IV.Eminence and Giftedness A.Intellect and Eminence 1. Eminence, renown, and fame 2.The real rewards of fame, renown, and eminence 3.The plight of the intellectual 4.Implications for the education of the gifted 5. Anti-intellectualism and eminence B. Research on Eminent Individuals 1. Biographical research on eminence 2.Psychological research on eminence V.Leadership and Giftedness A. Who Are Leaders? 1. Philosopher kings 2."The Prince' 3.The social compact B. Who Needs What Kind of Leaders?

106 115 Adulthood: Eitheetwe, Loadtrship, and Careen 107

1.Corporate leaders 2.Political leaders 3.Cultural leaders C.Characteristics of l.taders 1.Transactional and transformational leaders 2.Children's personality traits and leadership 3.Identifying potential adult leaders 4.The situational nature of leadership 5.The interaction between personalities and situations 6.Leadership and IQ 7.Are gifted children destined to be leaders? 8.Group dynamics and leadership training for gifted students 9.What is the program of study for transformational leaders? VI.Occupations and Giftedness A. Two Views of Careers B.Reasonable But Unrealistic Occupational Preferences 1.Fantasy 2.Career education 3.Professional status seeking and gifted children C.Giftedness and Professional Status 1. The IQs of professional groups 2.Careers among Terman's subjects 3.Conclusions and cautions D.Intellectualism and Careerism 1.The Stanford study 2.Implications E. Gifted Students, Intellect, and Occupational Life 1. Intellectualism and vocation 2.Careei:sm and anti-intellectualism 3.Occupational prowects for the gifted VII.Summary

Focusing Questions 1. Which of the four views of leadership most influences the content of leadership training programs for gifted students? Why? 2. What reasons might our society have for failing to cultivate or reward eminent scholars and artists? 3. What factors limit our ability to make generalizations from the research on eminent Persons? 4. How does the distincnon between transactional and transformational leadership relate to the provision of special instruction to gifted students? 5. What characteristics, other than high IQ, contribute to adult economic success? Does our society value these characteristics more than it values exr_eptional intelligence? Why or why not?

1 1 6 108 Adulthood: Eminence, Leadership, and Careers

CONCERN FOR THE ADULTHOOD OF GIFTED STUDENTS

This chapter investigates the subsequent adulthoodof children identified as gifted in school.After their public schooling, most of these children,but by no means all, complete four years of college.Many acquire professional degrees of one type or another, and a minority getdoctorates. What is likely to happen to them next? Shouldeducators try to ensure the future professional success of gifted students? Dochildren whom the school calls "gifted" really turn Out to be gifted adults?These are provocative questions, of course, and the adulthoods of ablestudents are inherently interesting to most teachers. In fact, many people believe thatgifteAl children will become our future leaders. They expect that able childrenwho have been well educated will provide the best possible nationalsecurity (e.g., Bull, 1985). Although this line of thinking seems reasonable, ithas not been confirmed by studies (Baird, 1985). Moreover, bothFeldman (1979) and Bull (1985) have highlighted weaknesses of this thinking.But the "greatest national resource" argument will probably be aroundfor a long time: Its truth seems self-evident to manypeople. You should remember, however, that goodresearch develops data that go beyond self-evidence. Teachers of thegifted need to examine the evidence of research so they can evaluate theplace of leadership training and career education in their programs.Looking at the research evidence can also help them examinetheir own self-evident assumptions about what will happen to their students. The issue ofadulthood is also important theoretically. Americans think that a prime purposeof education perhaps the main purpose of educationis to preparestudents for the "world of work" (e.g., Comminee forEconomic Development, 1985), and they often expect that gifted students will farebest in the job market (e.g., Hollingworth, 1926; Hoyt & Hebeler, 1974).Perhaps because we Ameri- cans maintain an optimisticview of human potential, we cherish great expectations for gifted students. What actuallyhappens when these students grow up, however, can also help usdistinguish between what education generally can and cannot do (cf. Katz,1971). The discussion below points out that definitionsof adult "success" often leave a lot to be desired. Hence, the discussiontries to clear up some of the confusion surroundMg theideas of eminence, leadership, and careers in general. Perhaps partof the difficulty, however, is that adult experiences are far more varied than theexperiences of schoolchildren. Success in adulthood may, therefore, be more commonthan success in school, whereas extreme success in aduhlaxyd maybe considerably more rare than childhoodgiftedness.

1 7 Adulthood: Esehieure, Leadership. and Careers 109

DEFINITIONS The key terms in this chapter are "leadership" and "eminence." Before proceeding, we need a brief look, at how the terms are typically used and the ways in which they are related.

Leadership Leadership has been investigated more fully than eminence as a topic of interest to teachers. The term, after all, was used to describe a type of giftedness by the federal government in 1970 (Kitano & Kirby, 1986). The endorsement of the federal government allowed schools to identify and operate special programs for schoolchildren who were believed to show special promise in leadership. Comparatively few schools provided programs, but the idea of special leadership skill still intrigues some educators, and it is frequently discussed in gifted education. The need to identify and train leaders seems to have arisen from the perception, shared by many people. that the solution of global pnthlems urgently requires leaders with a vision" (Kitano & Kirby. 1986, p. 251). This sense of the mission of education, however, is not new. It has been a cmicern of U.S. education since befOre Horace Mann (Cremin, 1980; Kati. 1968; Stevens & Wood, 1987). Its association with the gifted and its expression in special programs, however, is comparatively new. Starting in the 1920s, Terman promoted the notion that the gifted would provide a pool from whit.h future leaders would be selected by life. After 1968, when it became clear that Terman's claim could be contested (Feldman, 1984; cf. Oden, 1968), interest in programs to develop leadership quickened The most complete review of leadership and leadership training was completed by Stogdill (1974) and updated by Bass (1981). Siogdill's Hand- book organized theories of leadership into nineteen categories. The identification and training of nineteen kinds of leadership is probably not something for which the federal government was prepared to provide financial support, then or now. The scope of definitions of leadership runs from those that accord all authority to the individual to those that accord none to the individual. Some conceptions of leadership. in fact, assert that leaders make no difference (see Box 5-1).

Acceptable and unacceptable modds.Only if we can believe in individual leadership can we select individuals for special leadership training. If individual leadership is an illusion, however, then to identify and train it would be to practice deceit. Two of' the previous models avoid these cuhiesthe great man theory and the model of small group dynamics. Both give some importance to individual leadership. The great man theory is an attractive, if not completely plausible,

118 110 Adulthood: Eminence, Leadership, and Careers

BOX5-1Differences in Theories of Leadership

Rarely are the vast differences in theories of leadership madeclear in gifted education. The differences are so great that some styles of leadership are certain tobe at odds with prevailing political tendencies, whether right-wing or left-wing. Consensus onthe sort of leadership training to be provided is unlikely, given theseremarkable differences. Please observe that the differences discussed below representpolitical ideas, not par- tisan political practice. The Great Man Theory. The great man theory enjoyed the mostpopularity in the nineteenth century. According to this theory, a transcendent andcharismatic leader shapes major events. Certain types of biography and history recountpolitical and military leadership in the framework of this theory. The great mantheory of leadership is probably most consistent with a conservative interpretation of events.The reason is that the sphere of individual freedom is widest and the role of innate meritlargest in this conception. Great men, of course, need not, like Napoleon,be military or political leaders or, for that matter, even men. They could be artists (e.g.,Beethoven), philosophers (e.g., Karl Marx), social activists (e.g., Sojourner Truth), or nurses (e.g.,Florence Nightengale)all examples from the nineteenth century, please note, with equal representation to women and to left-leaning heroes. Small Group Dynamics. This view of leadership is ofmid-twentieth century origin. Its relevance pertains to professional life, whichoften requires us to take part in committees or work teams. Often such small groups dofunction poorly (e.g., Boss & McConkie, 1981; Mitchell, 1976; Ysseldyke, Algozzine, &Mitchell, 1982). Committee work does not entail the scope of mission that preoccupies 'greatpersons." The restricted scope reflects a narrower sphere of individualfreedom, and it assumes innate merit is a less significant influence on _vents. The modelof small group dynamics might be identified as the liberal view of lekership, A sizableliterature explaining how to influence such groups does exist, and skills can be taught togifted students (Feldhusen & Kolloff, 1979). The degree to which suchskills are effectively transferable has, however, been questioned (Woodman & Sherwood, 1980). The Group Attribution Theory. This theory of leadership,like the next, is a more critical interpretation of leadership than the preceding models.Leadership is auributed by a representative group. The leader is the spokesperson forthe group, but exercises no real independent authority. The leader's individual freedomand innate abilities are not really at issue, although the leader might possess somepersonal traits that enhance the leadership role. Possessing such traits would be convenient, but not necessary tothe role. The group attribution model might be called thedemocratic socialist interpretation of leadership. The Model of Institutional Legitimation. In this model the positionof leadership is structurally important to an institution, rather thanfunctionally important to a group that associates voluntarily. Leadership validates the institution, butthe leader can accomplish nothing at all by virtue of individual freedom or innateability. In fact, the entire group served by the leader operates with arestricted scope of freedom. In this model, the leader's authority derives from, andvia the misperceptions ofthose who are ledcontributes to the power of thelarger institution. This model might be interpreted as a totalitarian view of leadership. Adulthood: Erstagenee, Leadership, clad Careers 111 account of leadership. Part of its problem for education is that it does not provide a familiar setting in which to imagine the actions of a leader. In fact, we no longer create the kind of heroes described by the theory. The great man theory also fails to provide a set of definable skills that might be taught to potential leaders. The model of small group dynamics is both easier to contvive and easier to implement than a model of training based on the great man theory. For example, we know that there is a plausible connection between some gifted students and professional assignments of committee work. There is also a large literature on the topic. These features seem to make this model the natural choice for most gifted programs (Foster, 1981). Contemporary research, however, suggests that it is very difficult to improve the functioning of committees (Woodman & Sherwood, 1980) and that improvement may even produce undesirable institutional effects (Boss & McConkie, 1981). A program to train gifted leaders in the dynamics of small groups could be premature or even misguided. Leadership training would certainly be misguided if views of leadership, other than the small group dynamics view, were accepted as correct.

Identification.Not surprisingly, little has been done to develop adequate methods for identifying leadership ability as a specific talent. In addition to the unresolved questions about the nature of leadership. a major problem is the fact that leadership is a per-jut-mance skill. Like good teaching, leadership is difficult to train and cumbersome to measure. Moreover, adult leadership is different from the school leadership roles filled by students. Leadership training might best be provided by prokssional schools, where students roles more clowly approximate their work roles. Because methods of finding leadership potential are so primitive, the consensus of the field has been to provide leadership training to all students identified as gifted (Kitano & Kirby, 1986). Selection fOr such programs is, however, typically made on the basis of academic aptitude (Cox et al., 1985). Longitudinal studies suggest that these students are likely to enter occupations of higher status than unsekcted students. Whether or not they become leaders in their chosen fields, however, cannot he pre- dided from their academic aptitude (Baird. 1985). Leadri ship, howevet it is defined, seems to bear little relationship to intellectual ability (Bass, 1981).

Eminence The first use of this term in English, in 1420, corresponds closely to modern usage. "Eminent" denoted a quality "remarkable in degree" (Ox- ford English Dictionary, 1971, Vol. I). As used in the f011owing discussion, adult success is the quabty remarkable in degree.

1 0 112 Adulthood: Eminence, Leadership, said Carl% II Gallon (1869/1962) investigated eminence and developed ascale that measureo degree of remarkableattainment among adults. Eminence in Gallon's study represented extreme rarity of success. He rated thedegree of rarity of the success of his subjects upward to I in 1 million. Though Simonton (1984) has used similar techniques in hisstudies of historical figures. such ratings remain quite subjective. Part of theproblem lies in the nature of statistics. As a phenomenon becomes increasingly more rare, its description becomesincreasingly more prone to error. Thus, the confidence hand around a degree of rarity of 1 in 4 is small, but around a judgment of' 1 in I million it is quite large. Studies require impossiblylarge human samples to confirm the statistical reliability of a judgment as extreme in degree as 1 inI million. Galton's judgments of eminence were therefore quite unreliable by modern standards of statisticalreliability. Nonetheless, fbr all its statistical shortcomings, the idea of eminence is still intriguing. Why do some figures seem to exert extraordinaryinfluence on the thought and practice of theirtimes? These figures are our cultural heroes; their contributions to understanding and to the quality oflife are believed to be great; and their universal celebration provides de facto support fOr the great man theory. The prediction of eminence is also a motive. oddly enough.behind the identification of' gifted students (Bull, 1985; 11owley et aL.1986). It would be pleasant to learn that our cultural heroes were shapedin legitimate and meaningful ways by their experiencesin school. But because eminence is so rare, and because even our tests of academicaptitude and achievement cannot make reliable judgments as fine as 1inImillion, the accurate prediction of eminence from early promise isyietriumly impossible.

A continuum of talentOne way to view such qualities as academic talent, career success, kadership, eminence, and genius is as acontinuum from uncommon to extraordinarily rare. School "giftedness"is the lower end of' this hypothetical continuum of talent 'Thisdegree of rarityabout I in 50is a comparative rarity only among schoolchildren. Such rarityis in fact quite common among holders of ordinary occupations.Fewer than 1 person in 50 holds any particularjob. Bythis standard, fOr example. clerks, retail merchants, and educators are rare among the generalpopulation. If, however, one measures success as the initial qualification for a highly selectivecareertor example. graduating irom medical schoolthe rarity is somewhat more extreme, al,,,atIin 300 adults (Taylor. 1975). (The rarity of some occupations, of course, is not the result ofselection on academic talent, nor are all those who complete the training for ahighly selective career uncommonly talented in academics.) Career accomplishment as a physician is another matter still. Success might be measured by income, size of practice. or by contributions to

1 2 Aduhhood: Eminence. Leadership, and Careers 113 research or medical technique. Successful physicians will necessarily be even more rare than physicians. Among all physicians, whether successful or not, may be a number of physicians who are widely known regionally. nationally, or internationally for their work. They would constitute a group of renownd physicians. Among internationally renowned (or eminent) physicians might be a few recognized for medical genius. This small final group would constitute physicians who had made revolutionary contributions to medical knowledge and practice.

Studying the Relationships among Rare Phenomena It should be apparent that, as the dr-ges of rarity increases, it becomes more and more difficult to judge the quality by which accomplishment can be deemed successful, eminent, or of genius ("truly gifted") status. Only a few observers could even claim the right to judge the accomplishments of a hypothetical "medical genius." Such a judgment might be sound, but it might not be. Even eminent contemporaries often misjudge their "truly gifted" colleagues (Simonton, 1984). Although it may be impossible to determine the relative status of successful adults, we can nonetheless make some distinctions by analyzing the use of the terms "academic talent", "career success," "eminence." and "genius." The rest of the chapter will consider these distinctions in detail. It is, however, important to keep two fat ts in mind during the subsequent discussion:

I.Gifted adult ;icunnplishment is much nume rare than the phenomenon of exceptional academk potential. 2.The study of the relationship between the two phenomena is made difficult by their rarity and by the inaclequacv of the statistical tools used to study rare phemnnena.

Unwarranted claims characterize the development of programs for exceptionally talented students. The distinctions that f011ow should help teachers of the gifted distinguish between what special programs are likth to influence and what they are unlikriy to influence in the subsequent adulthoods of their students.

EMINENCE AND GIFTEDNESS

Gifted programs do not, of course, aim to make all uncommonly bright students eminent. Gifted education, however, has always valued the attempt to find and work with children who will eventually become eminent

1 ?2 114 Adulthood: Embassies, Leadership, and Carters (Gallagher, 1985; Terman, 1925). Mjst of usbelieve thatif we could locate themwe ought to provide suchchildren with a special program. One of the criticisms of traditional IQ testsis that they have failed to locate children who eventually become eminent(Feldman, 1984; Jenkins- Friedman, 1982). The critical questiondescribingprecisely the relationship between eminence and childhoodgiftednesshas not yet been addressed by much research, however. Part ofthe problem, of course, is that eminence is so rare, much more rare, infact, than childhood giftedness. Nonetheless, commitment to the search forchildren who are prospectively eminent adults has contributed to recentexperiments with new identification techniques, such as multiple selection measuresAnd biographical inventories. The relationship between childhcxxltalen,id eminence is an interesting topic for study, but to predict extremely rareadult accomplishment from any measure of childhood performanceis probably premature. It may even be an unwisegoal altogether (cf. Hersey, 1960). In any case, a great deal happens to peoplebetween the promise of childhood and their actual accomplishments as adults. If it is a problem to find children whowill subsequently become eminent, the problem of definingeminence in a way that educators can use is more striking still. Thedefinition varies not only from society to societybut from era to era within societies (Bull,1985). Of particular concern for educators of the gifted is the degree to which, at agiven time in history, what is popularly termed "eminence" is related tointellectual pursuits. Can we create a definition that accountsfin the fact that good luck (e.g., being the right color, coming from the rightbackground. and being in the right place at the right time) plays a role in whobeconws eminent? If not, then we are sure to repeat the sameabuses that have so often characterized the application of IQ tests.

Intellect and Eminence I'his chapter has already considered the htctthat intelligence does not priit eminence, defined as extremeadult accomplishment. The brightest Youngsters do not becomethe most eminent adults, evett thoughalxwe- average inte!ligemeprobably is required for eminence in mostfields (Baird, 1985; Aralbt.rg, Rasher, & Parkerson,1979). Conventional wisdom explains the inability of IQ tests to predictadult eminence as a Iv. 'mess of the tests (see e.g.. Alvino & Weiler, 1979).The previous discussi, .howed how the conventional wisdom really reflectsdifficulties associated with measuring extremely rare phenomena.This analysis was based on an appreciation of quantitative concepts. The f011owing discussim1 is ipuditative. Itexplains the lack of relationship between intelligence and eminence as aquestion of cultural values. It 123 Adulthood: Eminence, Leadership, and Careers 115 examines the way our values shape how we interpret the role of reason in society. The discussion explains two phenomena: (1) the frequent misin- terpretation of the observed low correlation between academic success and life success and (2) the failure of academic measures to correspond with measures of life success. Eminent persons embody our cultural values. Some of these values come from a cultural tradition of many centuries; others form the basis for new cultural legacies. For example, talented classical violinists become eminent because they help translate a long-standing cultural tradition of serious Western music. They perpetuate a heritage that is valued by an intellectual elite. Because works of serious music, art, and literature are valued solely by an intellectual elite, they represent "high culture." The key question here is whether or not the values of the intellectual elite determine the boundaries of' our cultural tradition. Can individuals who excel in fields other than tho.se validated by the high culture be considered eminent? These questions suggest the need to define such terms in a way that will enable us to make some useful distinctions.

Eminence, renown, and fame.Although the literature on gifted education does not qualify the term "eminence," it is enlightening to restrict the use of the term. In particular, one can distinguish among "emineme," "renown," and "fame." Although they are often used as synonyms, these three terms actually reflect relationships that exist between different groups of individuals and society as a whole. In common parlance, for example, one does not usually speak of an eminent football player. This sort of cultural hero is called "famous." So are rock stars, movie actors, popular singers, and certain corporate executives. Medical doctors who treat wealthy and famous clients are called mi- ther eminent nor fanmus. They might better be called "renowned." Fa- mous painters and theoretical physicists, may, however, be referred to as "eminent." especially if they are no longer living. The distinctions concern different levels of participation in the high culture by each category of hero; we call "eminent" only those heroes who thrther that legacy. Moreover, since the high culture represents a long historical tradition, it admits new fields quite slowly. Although science, for instance, has (xxupied increasingly larger portions of the intellectual legacy over the last three centuries, we still do not accord eminence too readily to living scientists. Within the high culture, however, rock music does not hold anywhere near the same status as science. It has not been around nearly so long. 'Therefore, we never accord eminence to rock stars, even posthumously. Classical violinists, on the other hand, are clearly involved in the high culture. When they become famous, they achieve eminence automatically.

?4 116 Adulthood: Eminence, Leadership, and Careen

The real rewards aflame, renown, and eminence.So far, this discussion might make it seem that the classical violinist is the most fortunate of the cultural heroes mentioned. At a higher level of abstraction, the discussion might also seem to indicate that the activities of the high culturewere the activities most valued in the United States. These conclusions, however, would be mistaken. For one thing, they misrepresent the way our society really works. Respect for the heroes of the high culture seldom involves the rewards that motivate most ofus and that most accurately represent the things we really value: substantial material rewards and effective influence in the larger society. Hence, the heroes of the high culturein spite of their eminencecontribute very little to the values we cherish most in our society. In fact, o r society, like most, has an ambivalent attitude toward both the high culture and the intellectuals who take an active part in it (Hofstad- ter, 1963; Perry. 1984; Shils, 1972). The intellectuals are by no means the ruling elite in our culture, nor do they promote the economic and political interests of the ruling elite (Gouldner. 1979; Holstadter, 1963; A. Howley, 1986; Shuts, 1972). Instead, most intellectuals reject or rebel against the ruling elitea trend that holds for other ancient and modern societies. whatever their economic or political systems (Upset 8c Dobson, 1972). The ruling elite is not interested in giving its heroes the status of eminence, but it can provide them with ample rewards, includingmoney, power, and renown. The group that warrants the status of cultural hero by the standards of the ruling elite is composed of techtuicratsapplied scientists, technicians, corporate executives, and university advisers to government and industry. Whereas some authors (e.g.. Perry, 1984) classify tech- tiocrats as a subgroup of intellectuals, the failure to distinguish between the two groups obscures the important mechanisms by which power and intellect are separated in our society ((;ouldner, 1979; Hofstadter. 1963. A. Howley, 1986; Mills, 1959). "1-he heroes of popular culture, thoogli they are not rewarded overth by the ruling elite are nevertheless controlled by that elite. Executives of the entertaimnent and communications industries haye almost exclusive authority over their fate (Mills, 1959). Popular stars help make huge profits for corporations, and their constant presence on TV and radio actually manifests the stability of our culture. A few popular entertainment and sports stars may take unpopular positions on public policy., hut their personal influence is necessarily limited by their econonr.: usefulness. Popular heroes also serve a useful role with respect to education: They provide support for the illusion that fame and the financial rewards of fame are accessible to anyone in our society. As a result,many students seek refuge in fantasies of fame in athletics and show businessfantasies that are as unattainable as they are self-deceptive.

125 Adulthood: Eminence, Leadership, aad Careers 117

The plight of the intellectual.Intellectual work in the United States is seldom valued for its own sake. It gets strong support, however, when it is used to further the goals of government, the military, and business. Hofstadter (1963, p. 25) distinguishes between intellect, which is "critical, creative, and contemplative" and intelligence, which is "manipulative, adjustive, [and] unfailingly practical." He characterizes America's reaction to intellect and intelligence (see p. 70). Because of the low value placed on intellectual work, U.S. society often does not accommodate intellectual vocations (Hofstadter, 1963). Sig- nificant intellectual activity frequently takes place outside formal economic institutions. Learned individuals often use their intelligence to direct careersas professors, lawyers, doctors, and executives. A small number of learned individuals use their intelket in activity that is best termed "avocational," since it is not profitable economically (Lopate, 1978). This intellectual activity may include scholarship, artistic production, and literary or political criticism. Even in universities, scholars have far le5s chance to do autonomous intellectual work than most people imagine.

Implications for the education of the gifted.Because there are very few avenues of employment that accommodate autonomous intellectual work, society, through the mechanism of the schools, guides intelligent youngsters away from scholarship for its own sake. Instead, it seeks to give them practice in the skills and, especially, the behaviors that characterize professional technocrats in our society. As a result, our schools do help to produce enough trained professionals and technical experts. However, they also teach the proposition that what is useful is of greater worth than what is beautiful or true. Schools do not seem to be cultivating intellectuals or artists. In fact, there is evidence that schools actually try to suppress rather than nurture intellect (Coleman, 1961; Oakes, 1985; Torrance ific Myers, 1971). Schools can actually function as an anti-intellectual force in society (Sawyer, 1988; Hofstadter, 1963). This may even be an importantif less than noble aim of schools with respect to their gifted students (Coleman, 1961; Tan- nenbaum. 1962), who have been found to cherish aesthetic and theoretical interests (Tidwell, 1980).

Anti-intellectualism and eminence.Certainly eminent attainment in intellectual and artistic spheres requires a high level of intelligence (see e.g., McNemar. 1964). According to Baird's (1985) carefUl review, the empirical evidence does suggest a positive relationship between intelligence and success of various forms. As Baird (1985) notes, the relationship may be greater than correlation studies indicate. Most such studies have involved only academically talentedoften giftedsubjects. The restricted

1"-'6 118 Adulthood: Emitirtee,Leadership, sued Careers range means thatobtained correlations are probably underestimates.De- spite the facts, popular opinion underestimatesthe relationship between academic ability and eminence. Why? For one thing, such a view helps supportthe belief that anyone can become eminent, a sentiment that reflects thenotion of fairness. Moreover. such a view justifies paying little attention tothe development of existing academic ability, which may be the resultof unfair advantage. Since the cultural institutions (e.g., symphonyorchestras and universities) cannot make mom for more eminent individualsthan they already manage to accommodate, the cultivationof many more such individuals would just increase competition for the fewplaces that do exist within these institutions. In general, our society functions asthough academic pursuits relate little to life success. Thus our social institutions(particularly schools) cultivate noncognitive behaviorsthat have a more practical connection tolife success (Baird, 1985; Bowles& Gintis, 1976; Hoistadter, 1964; A.Howley, 1986). These social institutions conditionbehaviors that support the continued prosperity of business, industry, and government.In the process, however, they alienate those intellectuals who mayprovide access to an importam past and future legacy. Research on Eminent Individuals There has been very little research on thecharacteristics of eminent individuals. In part, research has been limitedby the extreme rarity of the kinds of accomplishments associated witheminent status in the high culture. Researchers find itdifficult to identify a large sample of eminent individuals at any given time. Because of thisdifficulty, some researchers (e.g.. Cox, 1926; Goerrzel & Goertzel,1962; Simonton, 1984) have analyzed historical reports of eminent personsfi-om many eras. This research is based on a narrow definition of eminence,similar to the one that we have articulated. Other research (e.g., Segal. Busse, &Mansfield. 1980) is based on a broader definitionof Aiperior performance. and the results ofthis research will be considered subsequently,in the section of this chapter devoted to leadership. Biographical research on eminence.Most of the studies about eminence are based onbiographical records of men and women whohave earned national or international reputationsfor accomplishment in scholarship, politics, or the arts. Some findings arebased on studies of contemporary eminent subjectsand include the results of interviews, aptitude tests, and personality measures.Most of the findings characterize the nature of eminent men and womenand their families. A few studies have considered broader influences, such as geographyand popular culture. In general, the biographical research oneminence is more circular, 127 Adulthood: Eminence, Leadership, and Careers 119 more anecdotal, and more speculative than most contemporary research in psychology and sociology. Seldom do the biographical reports of eminent persons contain data that can be manipulated statistically. The few quantitative data that do exist vary among subjects. Moreover, since there is so much biographical detail to choose from, selection and interpretation are largely a function of the researcher's bias. The empirical literature on eminence began with Galton's (1869/ 1962) Hereditaly Genius, which reports his assumptions, methods, and conclusions about eminent men. His conclusions led him to prescribe eugenics as the means of improving human achievement, since he believed that heredity alone was sufficient to explain eminence. His conviction that heredity was virtually the only source of resemblance among members of the families of eminent persons was completely unwarranted. A distinctive family environment was as likely a source, and few researchers since his time have assumed a position so extreme. Some have taken the opposite view; most have taken an intermediate position. Odin (1895), who studied eminent French literati, disagreed sharply with Galton's conclusion that no man of innate exceptional ability could be deterred from eminence by environmental circumstance. Ward (1906) presented the viewpoint that ascribed significant influence primarily to environmental forces. In Applied Sociology, Ward (1906) asserted that the working class is as capable as the upper classes of producing eminence and that the difference between the classes lies not in innate talent but in privilege. Clarke (1916/1968), who was influenced by Odin and Ward, included demographic variables in his study of American men of' letters and found that several geographic and economic variables were associated with literary eminence. Taking population size differences into account, he found that the greatest proportion of literary men and women in his sample of 1,000 came from the New England states. Tlw southern and northwestern states contributed the fewest subjects. Thirty-two percent of the American literati were from relatively metropolitan areas even though these areas contained only 9 percent of' the population at the time Clarke's subjects lived (none was bc,rn later than 1850). Seventy-eight percent came from families with at least average incomes. According to his statistics, a man or woman with some college education was several hundred times inoi e likely to achieve literary distinction than someone without any college education. Catherine Cox (1926) was concerned primarily with the psychological characteristics of eminent persons, particularly their degree of' intellectual ability. She attempted to estimate childhood and adult IQ scores for 301 famous men and women based On juvenilia and biographical records; her subjects included some of the most famous figures in English, European, and American history. The group with the highest estimated childhood IQs, according to Cox's formula, was composed of the 12 philosophers 1 r',JS 120 Ads Mood: Enthsenee, Leculenhip, and Careers among her subjects. Their average estimated IQ was not less than 175. Cox suggested that a "true" IQ (i.e., actually measured at the time the subjects were children) for the entire group would not have been lower than 155 to 165. Cox's estimated data, however, yielded an average IQ about 20 points lower. On the basis of her study. Cox characterized these subjects as having above-average heredity and superior advantages in eaily environment. Davis (1929) studied the life histories of 163 leaders of the Russian revolution. On the basis of historical and biographical records, he concluded that the greatest influences on these eminent men were teachers, revolutionary literature, and the idealism of their father or mother. lie also noted that the leaders had nearly all done university work and that most of them came from middle- and upper-class families. In Cradles of Eminence, Goertzel and Goertzel (1962) described the emotional and intellectual climate in the homes of eminent persons. 'they included in their study 400 Americans and Europeans who lived into the twentieth century. They concluded that virtually every subject had at least one parent who showed a high degree of commitment, ability, or emotional intensity. As a result, family life was neither pleasant nor comfortable; it was often tumultuous. Their research led Goertzel and Goertzel to suspect that eminence develops in families that cultivate extreme levels of intellect, emotion, and commitment to principles. Psychological research on eminence.Rese:sch on eminent contemporaries has been conducted by psychologists interested in creative behavior (see Chapter 4 for research related to creativity per se.) Recent studies have focused on the effects of intellectual abilities, persoaality traits, life history, and family background on the emergence of exceptional talent. Like earlier studies, they found that eminent persons have exceptional ability. In most cases, although their talent in early childhood was evident to their families, it was not evident to their schools. Recent studies of eminent individuals have therefore tried to identify personal and familial characteristics that condition eminence. Economic and social influences have been examined less closely, though they are equally important (cE Jencks et al., 1979). Like earlier researchers, contemporary writers have also noted that eminent persons and their parents are committed to learning and to principled action. Bkxml (192), in particular, in his study of internationally recognized young mathematicians, athletes, and musicians, called attention in the extraordinary commitment of Emily time and resources necessary for superior achievement.

LEADERSHIP AND GIFTEDNESS Akin to the notion of eminence, leadership implicates certain kinds of exceptional performance. Unlike eminence, however, leadership has been considered extensively ffom both philosophical and empirical perspectives. 1 29 Adulthood; Eisisseitee, Leadership, and Careers 121

This brief survey of the empirical literature on leadership first presents an analysis of the idea of leadership; then it presents research findings about leadership. Finally, it applies both theory and research to an analysis of the role of leadership training in the education of gifted students.

Who Are Leaders? Political philosophy has considered, throughout history, the wattle and reie of leadership. Philosophers have articulateu ."fterent views of leadership, based on their more general notions about the structure and function of government. In their consideration of leadership, some philosophers have also described the personal traits and types of behavior required of a leader in an ideal system of government. Our discussion considers three philosophical views of leadership and characterizes each view by briefly examining the work of one representative philosopher. A meritocratic view is presented in Plato's description of the philosopher king in the Republic.. An autocratic view is expressed in The Prim-e by Machiavelli, and a democratic view is expounded by John Locke in Two Treatises of Gwernment.

Philosopher kings.In order to understand Plato's political philosophy, it is important to know something about his metaphysics, the branch of philosophy that considers questions about the ultimate meaning of life. For Plato ultimate meaning involved knowledge of ideal reality (i.e., the way things really are as opposed to the way things seem). The purpose of each individual's life and the collective life of individuals in society was to conform to the ideal and thereby to achieve goodness. Because, for Plato, knowledge of the ideal was the essence of life, he saw as natural leaders those individuals who had the greatest access to this knowledge. He maintained that knowledge of the good was equivalent to goodness and that evil was merely ignorance of the good. Since government existed to rid society of evil, those who governed should be those most knowledgeable of the good. By this logic. Plato concluded that philosophers were the most capable rulers. Plato compared the leadership of philosophers with that of politicians and intellectuals. He fOund politicians to be concerned with gaining and keeping political power. an activity that entailed evil actions. He found intellectuals to be totally unconcerned with action and therefore apolitical. Philosophers, according to Plato, were able to resist the allure of power while still being able to enact the good. This combination of virtues resulted from their inherent abilities and an appropriate education. Plato maintained that leaders (whom he called guardians) needed to have physical strength, courage, and a philosophic temperament. These inherent capabilities were not sufficient, however, to produce philosophers capable of ruling. In addition to these innate capabilities, potential leaders 130 122 Adulthood: Eminence, Leadership, mod Careers II needed extensive education. Their education involved additional studies beyond the primary and secondary education that Plato advocated for all citizens. Thus, philosophers needed to receive primary education that involved reading, writing, and physical education; secondary education that involved the study of literature and music; and higher education that involved the study of mathematics and pure philosophy. Plato viewed government as rule by an intellectually and morally superior elite. He understood that such a form of government would be unpopular with the citizenry. The role of the leader, however, would be to provide the citizens with knowledge of the good. When citizens were aware of the good, they would be less likely to resent the rule of the philosopher kings.

"The Prince."In contrast to Plato, Machiavelli was concerned with how leaders function in the real (as opposed to the ideal) world. He wrote:

How we live is so far removed from how we ought to live, that he who abandons what is done fOr what ought to be done, will rather learn to bring about his own ruin than his preservation. (Machiavelli, reprinted in Harris, Morgenbesser, Rothschild, & Wishy, 1960, p. 469)

Although Machiavelli wishedit were otherwise, he nonetheless viewed humankind as disorderly and evil. He believed that humans needed to be governed by powerful leaders whose rulewould provide order and stability. Based on the premise that any government was better than chaos, Machiavelli set out to define the method by which rulers could get and maintain power. Machiavelli's political philosophy was founded on the assumption that the goal of establishing order justified any means of getting and retaining power. He believed that fewer people would suffer as aresult of a prince's rise to power than would suffer as a result of anarchy. Hence. Machiavelli believed that it was ethical to recommend that rulers seize and maintain power by force or deceit. He wrote:

A man who wishes to make a profession of goodness in everything must necessarily come to grief among so many who are not good. Therefore it is necessary for a prince who wishes to maintain himself, tolearn .;ow not to be good, and to use this knowledge and not use it, according to the necessity of the case. (Machiavelli, reprinted in Harris et al.. 1960, pp. 469-470)

According to Machiavelli, the prince should learn to make use of whatever virtues and vices would be required in order to maintain power. To learn how to govern in this way, the prince needed an education in the organization and discipline of war. This study entailed the practice of military maneuvers even during peacetime. Such practice was intended to n Adulthood: Eminence, Leadership, and Careen 123 accustom the prince to physical hardship and to acquaint him withthe terrain of his principality. This knowledgewas necessary to assure the prince's military success in thewars that would be an inevitable part of his maintenance of power. In additionto physical training, Machiavelli recommended that the prince study historyso that he might understand the ways in which leaders from othereras gained and used power. Like the rulers that Plato envisaged, Machiavelli'sprinces were not likely to be popular. In fact, Machiavelli suggestedthat it was better for them to be feared than to be loved. He warned,however, of the dangers of being a hated ruler. He believed thata prince should keep from being hated by refraining from stealing his subjects'property. He also suggested that a prince might improve his popularity bypromising to bestow various benefits on his subjects. Machiavelli, however, didnot require that the prince actually deliver promised benefits.

The social compact. The philosophy of JohnLocke implied that political leaders should be very different from thesorts of rulers put forward by Plato or Machiavelli. Locke's view of' thesocial compact could be practiced only if leaders represented the interests ofthe public. Leaders, therefore, needed to bevery much like the citizens whom they represented. Locke's views were based on two premises: (I) thatthere are certain natural rights accruing to all people and (2) thatgroups of people can willingly give up some of their natural rights in orderto be protected by constitutional rule. Government, accordingto Locke, should be based on the principles of equality and liberty. Eachindividual is equal because each has the same natural rights. Each individual is freeto exchange some of these natural rights for the protection ofgovernment. Such a government, however, would be legitimateonly if it func- tioned to protect the interests of the majority of' itscitizens. Its legitimacy also would depend on the fact that it allowed peoplewhose interests were Hot represented to withdraw from the compact and seek other,more favorable governments. In addition, sucha government would be effective only insofar as its structure allowed for theseparation of the law-making and the law-enforcing functions. Thus, the democraticgovernment advocated by Locke needed to be composed ofa legislative and an executive branch. Locke suggested that a third branch ofgovernment, the federative branch, should be responsible for international affairs. Because the leaders of governmentwere accountable to the majoi ity of citizens, Locke did notconcern himself with their characteristicsor training. His approach to democratic leadershipwas based on the assumption that all citizens would be equally capable ofserving as political leaders. This view of leadersLip strongly influenced themen who framed the U.S. Constitution, and it even lent support (a hundredyears later) to those who advocated the compulsory education of all children.

132 124 Adulthood: Eminence, Leadership, and Carters

Who Needs What Kind of Leaders? It is clear from the discussion above that the nature and role of good political leaders depends on the qualities of the government that they serve. Common sense suggests that complex societies may need many sorts of leaders and that the characteristics of these different leaders may vary according to the nature of the enterprises they lead. Because of the diverse roles filled by different sorts of leaders, it may be hard for society to predict the characteristics of potential leaders (Bull, 1985). Nevertheless, the rationale for providing leadership training to gifted students is that such training will help them in whatever leadership positions they eventually find themselves (Foster, 1981). The discussion that follows evaluates the nature of leadership in contemporary social institutions. It compares the types of roles assumed by corporate leaders, political leaders, and cultural leaders. It also tries to determine the degree to which particular personality characteristics or social behaviors assure the success of' diffrrent kinds of leaders.

Corporate leaders.Although the prototypical American businessman is the entrepreneur or "self-made man", this type of businessman is neither the num powerful nor the most prevalent in modern U.S. society (Mills, 1959). The most influential corporate executives work fOr companies that are owned by wealthy individuals or families, some of whom have maintained wealth over many generations and others of whom have acquired vast sums of money more recently. The leadership role assumed by these owners varies greatly; such owners often assume no leadership role at all. In the most common case, corpoi ate leadership comes from managers rather than the founders or owners of corporations. According to Bass (1981), there are innumerable typologies of leadership. Many of these typologies are designed to characterize corporate managers. Bass (1981. P. 22) suggests that the f011owing types of leaders are common to many leadership typologies:

auttmritative dominative directive auuKratic persuasive arousing charismatic secluctive convincing democratic participative group developing considerate intellectual eminent expert executive bureaucrat administrator bead representative spokesperson advocate

He also notes that "most recent researchers have devoted comparatively little attention to three of these types: persuasive, intelfrctual. and representative" (Bass, 1981. p. 22). Adulthood Elegance, Leadership, amid Careers 125

There seems to be no consensus aboutwhat type of leadership is most appropriate for a particular managementfunction. Bass (1981) suggests that to understand effective management onemay need to analyzethe system in which theleader is embedded. This analysiswould include the study of "multiple inputs fromthe environment, the organization,the immediate work group supervised,the task, the leader's behavior,and relationships with subordinates, and outputsin terms of effective performance and satisfactions"(Bass. 1981, p. 613). Clearly, no one style or method of managementdetermines success as a corporateleader. 'Therefore, it would be difficult toidentify potential corporate leaders on thebasis of their personal characteristics or tobase leadership training for corporate success on anyparticular set of behaviors.

Political leaders.While many of' the characteristir- political leaders are conditionedby the requirements of their role ,e media and the public relations industry also influencepolitical leadership (see e.g., The Ripon Society, 1969). Training inparticular leadership skills may play a part in political leaders'exercise qf power. but it probably does notinfluence their ability to govern wisely or to representthe commonweal. According to Mills (1959), there are two typesof political lead,...rs, those who assume executive functionsand those who assume legislative functions. Of these two types. Millsfinds the executive leaders to be more powerful because of their close connectionswith the corporate and military elites. The success of executive leadersdepends on their ability to make use of connections with other powerful groupsrather than on their ability to develop or exploit particular personalqualities. Mills believes that legislative leadersdo not have any real authority to represent the majorityof citizens. First, he asserts that themajority of' citizens do not constitute an informedpublic that guides the actions of its representatives. Second, Mills suggeststhat any legislative representative must secure and maintainthe backing of the powerful constituentsin his or her district. Thus, legislators arebound to develop and vote for laws that promote the interests of their mostpowerful supporters. According to Mills, legislators are the middlemenwho promote the interests of the ruling elite. Like Machiavelli's princes,political leaders may need to act in ways that serve to keep them in power.Although the methods of maintaining power need not entailvirtue, they cannot be too overtly dominatedby vice (Mills, 1959). Nonetheless, according toMills, our society has traditionally tolerated a considerable amountof wrongdoing on the part of its political representatives. Because of their close affiliation withthe ruling elite, most political leaders are better educated than the averagecitizen. Many of them are lawyers or businessmen. In spiteof their educational attainment, however,

134 126 Adulthood: Enhance, Leadership, and Careers

most political leaders disassociate themselves from intellectuals and deni- grate scholarly pursuits (Hofstadter, 1963). The anti-intellectual posture assumed by political leaders serves to safeguard their existing popularity and to protect them from the appearance of elitism (Mil ls, 1959).

Cultured leaders.The characteristics of and the roles assumed by the most notable cultural leaders were considered in the preceding section on eminence. In that discussion, eminent cultural leaders were viewed in terms of their ability to advance the high culture. The production of enlightening works of scholarship or beautiful works of art hardly ensures eminence, however. Many cultural leaders are not, in fact, eminent. Some cultural leaders may remain unrecognized in their own eras but later achieve eminent status, but many more live and work in relative obscurity. To prepare anyone for a leadership role of such a speculative nature is clearly impossible (Bull, 1985). On the other hand, it may be possible to provide an education that acquaints many capable students with significant cultural traditions. By educating many students in these traditions, we can perhaps simultaneously nurture those few who will become cultural leaders (Howley et al., 1986). Because cultural leaders are nmst often intellectuals or artists, their interests may be contrary to the interests of imlitical and corporate leaders. Anti-intellectual forces in our society may make it difficult for many cultural leaders to wain popular support or economic reward (Hofstadter, 1963; Upset & Dobson, 1972). Cultural leaders may actually exist in spite of our sGcialinstitutions(including schools)rather than because of them lastadter, 1963). As a group, cultural leaders are probably even more diverse than the corporate or political leaders discussed alxwe. It is very unlikely, therefore, that the sorts of leadership training that might provicie some benefit to corporate or political leaders would have any bearing on the contributions made by cultural leaders.

Characteristics of Leaders A great deal of research has aimed at identifying traits that characterize leaders. Some of this research investigates the characteristics and behaviors of capable managers, whereas other research explores the traits of charismatic politicians. It is possible to interpret the research only when it is organized around a theoretical premise. Genrralization, however, is very difficult because the research is based on so many different definitions of leadership.

Transactional and transformational kaders.Burns (1978) makes a theoretical distinction that allows for a meanMgful synthesis of the research

135 Adulthood Eminence, Leadership, sued Comers 127 on leadership. He suggeststhat there are two distinct typesof leaders, transactional leaders and transformationalleaders. These types of leaders serve very different sortsof social functions and consequentlybehave quite differently. According to Burns (1978, p. 19),"transactional leadership ..occurs when one person takes the initiative inmaking contact with others forthe purpose of anexchange of valued things."Transformational leadership occurs, on the otherhand, "when one or more persons engagewith others in such a way that leaders andfollowers raise one another tohigher levels of motivation and morality" (Burns,1978, p. 20). Burns classifies managers as transactionalleaders and charismatic politicians astransformational leaders. In the discussimi that follows, wewill use Burns' distinction tohelp explain some of the researchfindings on leadership. In particular,his distinction will help us evaluate thedegree to which leadership can, or even should, be trained. Oildren's personality traits andleadership.Researchers have attributed a wide variety of personalitytraits to leaders, includingsuperior intelligence, motivation to excel,originality, judgment, sociability, aggressiveness, generosity, dependability,scholarship, and physical prowess (Stogdill, 1974). Various studies havealso fOund personality differencesin leaders on the basis of the sex,ethnic membership. socioeconomic status, and other characteristics oftheir constituencies. Sex-relateddifferences associated with leadership have beenfound among children insexually homogeneous and heterogeneous groups.In elementary school, boys' leadership seems to depend on verbalaggressiveness and girls' leadership seems to depend on success atschool tasks (Zander & Van Egmond,1958). Male leaders of boys' gangs andplay groups are usually above averagein athletic ability and physical prowess(Partridge, 1934). An extensive study of boys' gangs in Chicago foundthe salient trait of the leader ofthe group to be his gameness(Thrasher. 1927). In girls' dot mitoriesleadership is correlated with generosity, enthusiasm,and affection; dorm leaders tend to be characterized asprotective of the weak, dependable,tactful, considerate, confidence-Mspiring,and able to establish rapport quickly(Jennings, 1943). In general, leaders tend toepitomize the values of their [(Mowers (Litzinger & Schaefer, 1984). Superiorsocioeconomic status and scholastic standing, for example, differentiate campusleaders from other faudents (Sward, 1933). The most influentialcollege campus members represent dominant values on the campus (Stogdill,1974). These findings suggest that an acceptable leader mustinternalize a group's norms (I-itzinger & Schafer, 1984). Group values, however, are subject tochange. For example, student

1 3 6 1111 Ashdthaack Ensimisee, Leadership, and Careen Ieaders of the 1960s differed from .udent leaders of other periods.The unpopularity on college campuses of the Vietnam War contributed to the emergence of student leaders who repudiatedestablished political and economic beliefs. Leadership, then, is not merely the possession of a combination of personality traits. An individual may be able to rise to the top with one group of peers but not with another; an individual who is aleader in one era may not be an acceptable leader in another era (Stogdill,1974).

Identifying potential adult leaders. The identification of children who might become leaders as adults often proceeds from a view of leadership based on personality traits. This view also influences the contentof leadership training programs. The validity of predicting children's future leadership from studies of the personality traits of adult leaders has not, however, been established. Moreover, it is not even clear that the general personality traits of some adults condition their performance as leaders. Much of the research on leaders' personality traits is based on subjective accounts. Iii the first half of this century, authors biased their accounts of leaders' characteristics by seeking to identify predetermined qualities (Bavelas, 1984). More recent research efforts have been limited by thelack of an adequately standardized instrument to measure general personality traits that predict leadership capability. Recent research, therefore, has relied on experimental instruments for identifying leadership potential among specific groups and in specific situations.

The situational nature of leadership.According to Stogdill (1974), to understand leadership, one must analyze situations as well as leaders. Power is an important determinant of situational leadership. Sources of power include expertise, legitimation,coercion, reward, and reference (liking) (Stogdill, 1974). Since personality traits may help explain why individuals are liked, they may influence effective transactional leadership to some degree. If not assigned by superiors, committee leaders are those who assume ad hoc responsibility for seeing that the committee accomplishes its task. They are not necessarily more competent or more popular than other committee members, although they seem to participate more in group discussion (Stogdill, 1974). Committees often cede leadership to their most active member. In these cases, committee members are willing to relinquish control of the task because they believe either that all committee members are equally competent to accomplish thetask or that the commatee's task is not very important. Research simulations of committee work seem to reproduce the conditions under which some committees actually function. The committees to which these findings are applicable are those in which the stakes for performance are not high, committee members are nothostile to one another,

137 Adulthood: Eminence, Ltoderthip, and Carver: 129 and the task is imposed by a benign external authority. Cooperation with a particular group leader under such circumstances is more likel). to depend on the leader's likeableness than on other sources of power. Effective transactional leaders in experimental groupsand in the actual groups they simulatetend to be spontaneous, tolerant of a range of ideas, accepting of diverse personalities in the group, and solicitous of participation by all members. It is, however, clear that many real life committees function under circumstances quite different from these experimental conditiors. For example, an employee who is not liked may be designated to chair a committee. In this case, the preceding discussion does not apply. For transformational leadership, where the stakes are higher, personality traits probably carry far less weight than other sources of power, such as political or economic leverage. Many great transformational leaders were isolates; many were irascible, tactless, and overbearing. Interpersonal skills are relevant to effective leadership primarily when the leader's main source of power is likeableness. Personality traits should not, however, be expected to predict leadership in situations where significant cultural, political, or economic change is at issue.

The interaction between personalitiesand situations.According to Bavelas (1984), a leader exhibits leadership in some situations, but not in others. The degree of structure, the cooperativeness of the group, the nature of the task, and many other variables determine the personality traits required for effective leadership. The individual who is at ease and capable of influencing a group under certain circumstances may not be effective under other circumstances. Authoritarian leaders, for example lf e more likely to accomplish clearly defined, important goals; they are le. to be effective in situations where goals are unclear or unimporti. the group's well-being (Konen, 1972). On comparatively rare occasions, the effective leader's personality may affect the context in which transactional leadership occurs. Different leaders, for example, define and structure tasks in different ways, in part because of differences in personality. Such differences of style may, in certain situations, distinguish between merely competent and noticeably ad-oit leadership. When transactional leaders significantly affect the context in which their leadership occurs, however, they are behaving like transformational leaders. This metamorphosis may be appropriate; on the other hand, it may exceed the requirements of the organization in which the committee exists. Success or failure to accomplish limited tasks depends, in part, on com- paibility between the leader's personality, the nature of the task, and the composition of the group. Committees that become too effective can cause organizational problems (Boss & McConkie, 1981).

138 130 Adulthood: Eminence, Leadership, and Careers Leadership and IQ.Leadership is included in the federal definition of giftedness in Mar land's (1972) report to Congress on the status of gifted education in the United States and in the revised definition published in Section 902 of PL 95-561, The Gifted and Talented Children's Education Act of 1978. Of the five types of giftedness established by the definition, leadership is the least frequently addressed through identification efforts. As we have seen, research offers no good reason for labeling a particular group of children as uniquely gifted inleadership. Educators lack valid and reliable means of identifying the potential for leadership in young students. It is not possible to exclude any child from a leadership program on the grounds that the child lacks leadership potential. Concern for leadership talent is therefore more commonly addressed through leadership training provided to intellectually gifted children. The justification for this approach is that the research consistently reports that lcaders are above the average intelligence level of the group they lead. In addition, many educators cite the correspondence between some of the personality traits of leaders and some of the traits of high-1Q children. Clark (1979), for example, believes that leadership entails skills that enable a grovp to maintain itself, reach its goals, adapt to environmental change, and permit self-fulfillment among group members. According to Clark (1979, p. 236), "leadership skills are mostly interpersonal and include flexibility, openness, and ability to organize. Leadership requires self-esteem, high values, and mature emotional development."Clark believes that these characteristics are compatible with traits typical of gifted children. It is probably true that many high-ability students and few low-ability students possess the cognitive skills needed for leadership (Gallagher, 1985). "High ability" in this case does not mean exceptional ability, however. Research indicates that the leader is usually only slightly more able than the group of followers; the correlation betweenability and leadership is not very strong. Zander and Van Egmond (1958)found a low positive correlation between intelligence and social power among second- and fifth-grade students. Their correlation is about the same as the average correlation of other studies, approximately 0,28.

Are gifted children destined to be leaders?In general, leaders appear to be above the group average, but not too farabove it. The individual with an IQ of 160 is not likely to be the leaderof an average group. According to Hollingworth (1926), an IQ range from 115 to 130 is more typical of leaders of average-IQ groups. Children with !Qs of 160 a id above are not excluded from leadership of pfted groups, however (Hollingworth, 1926). Subsequent studies have confirmed Hollingworth's assertions. A study of managers found that individuals with either low or very high scores were less likely to achieve success in managementpositions than Adulthood: Eminence, Leadership, cud Canters 131 those with scores at the intermediatelevel (Ghiselli, 1963). Other studiesof businessmen report the same finding(Baird, 1985). In contrast to the expectation that gifted children wouldbe natural leaders, it may be likely that many gifted children are less apt tobe leaders than children with IQ scores only slightly above average. This observation may be true notonly of high-IQ children, butof children with superior divergentthinking ability as well. In a study of elementary children, Torrance (1963)found that the most highly divergent thinkers among theolder children were often ignoredand subdued by the other members of the group.Torrance (1963) was surprised to find that the older and more organizedthe group, the more it inhibitedthe member who had sccred highest on a testof divergent thinking. He com- mented on this effeq:

Plainly, the tendency of- organization tocontrol the most creative tends to coerce some highly creativechildren and reduce their usefulness. There are, of course, the aggressive, irrepressiblesubjects. .There are also diplomatic, creative persons who kel their way graduallyand slowly win acceptance of expend so much their ideas ...That creative people should be required to energy in "being nice" inorder to obtain a hearing fOr their ideasis regrett- able. (lOrrance, 1963. p. 135)

Although 68 percent of the top scorers ondivergent thinking tests initiated more ideas than any oth et. membersof the group, only one-fourth of these students were credited asmaking valuable contributions to the group's performance. Only the mostpersistent or nmst diplomatic of the top scorers were actuallyable to make a contribution; otherswithdrew from the process, disrupted it. oraccepted minor tasks that denied them the opportunity to contribute. If leaders tend to epitomize the valuesof the groups they lead, the interests and behavior of intellectuallygifted persons may inhibit their efiectiveness as leaders. Opportunities forleadership by the intellectually gifted are most likely to arise as aresult of their expertise. They might, through professional influeme, lead groupsconstituted of other capable persons in their fields.Gifted persons might also occupyleadership positions as a result of characteristics obscurelyrelated to their giftedness, such as attractiveness oreconomic power.

Group dynamics and leadership trainingfor gifted students.Research on group dynamicstypically suggests the content forleadership training programs for giftedchildren. Such research investigates transactionallead- ership exclusively; programs based onresearch about group dynamics typically try to develop communicationskills, decision-making skills, planning skills, and skills in values clarification (Karnes& Chauvin, 1986).

140 132 Adsdthood: Enshmsee, Leadership, and Careers One such program described by Foster (1981)trains adolescents who have been identified as gifted in social ability. The program attemptsto teach group communication skills, problem solving, anddecision-making procedures. Students are matched with mentors whohold leadership positions in local political, business, and social organizations.Since intellectual giftedness is not a prerequisite for success in these programs,it is unwise and unfair to exclude average ability students who couldbenefit from them (Cox et al., 1985). Such programs include limitedopportunities to improve academic skills, but they are unlikely to providesufficient academic challenges to intellectually gifted students.

What is the /Program of study for transformationalleaders? When educators discuss the need for leaders who can solveimportant national and international problems (e.g., Cox et al., 1985;Torrance, 1984), they are clearly referring totransformational leaders. Tiansformational leaders assume responsibility forwork that is both more dangerous and more important than the committee work of transactionalleaders. In providing transactional leadership training to intellectually giftedchildren as a means of cultivating leadership potential, educators maybe making two serious errors. First, they may beproviding the wrong sort of training to a talented group. Second, by devaluingintellectual skills, they may discourage children of all ability levels from acquiring the knowledgeand skills that will enable them to understand and participate in significantpolitical, economic, and cultural changes (cf. Co lemon, 1961: Katz, 1971;Scott, 1986). In a comparison of charismatic and consensusleaders, Zaleznik (1984) proposes that major change.- is brought about primarilyby leaders who do not conform to accepted values in the area in which theyeffect change. If leaders are to be creative, they may have toviolate accepted standards of polite behavior (McCall, 1984). The social valuesand transactional skills taught to gifted students in leadership training p; ograms may therefore actually hinder the development of transforniationolleaders. The potential leaders of significant artistic,scientific, or political movements are better served bytraming that develops superior intellectual competence. Transformational leadershipusually requires an understanding of the history of the pertinent discipline and therelationship of' its principal concepts and methods to other disciplines (cf. Kitano& Kirby, 1986). The power that legitimates transformational leaders, asnoted previously, is not likely to be interpersonal skill. Programs forintellectually gifled students should cultivate cognitive skills; teachingnoncognitive skills is unlikely to prepare intellectually gifted students to exercisethe sorts of leadership roles for which they do seem to be best suited. On the contrary,it may well inhibit them in suchroles.

141 Adulthood: Eminence, Leadmhip, and Careers 133

OCCUPATIONS AND GIFTEDNESS

Virtually all men, and increasing numbers of' women,take paid occupations after they finish school. The objectof this portion of the chapter is to provide a background for understandingthe characteristic relationships of giftedness (academic and intellectualtalent) to the performance of gifted individuals in occupations and careers.Teachers of the gifted need to consider these matters because career preparationprogramming for their students is increasingly recommended (e.g.,Feldhusen & Kolloff, 1979; Fleming, 1985; Van Tassel-Baska, 1981).

Two Views of Careers In a broad sense, a "career" is anindividual's occupational progress over an entire lifetime, and anoccupation is the general activity to which an individual devotes the most time and effort at agiven time. In one view, neither occupations nor careers need be paidemployment (Fleming, 1985; Hoyt & Hebeler, 1974). For example, a personwho raises five children, without ever receiving a salary or wages for theeffort, can be said to have a career. The problem withthis definition, however, is that it promotes a view of careers that is blind to gender. race,and class prejudices. It ignores factors that strongly influence the occupationalfate of everyone (Wright, 1979). Perhaps this omission is necessary in careereducation in the interest of equality of opportunity, but it is not veryhelpful in clarifying the characteristic occupational progress of giftedindividuals. Ginzberg (1971) uses the term "career"differently in his survey of' career guidance:

Since niost blue- and white-collar workersseldom move more than a few rungs up the skill and incomeladder during the course of their working lives, their work experience, with its limited progression, cannotbe called a career. (Ginzberg, 1971, p. 7)

Ginzberg's definition better approximates the common senseof "career." It also recognizes such influences as elevatedsocial status, high income, and other labor market influences (such as gender, race,and social class). For three important reasons. thesubsequent discussion follows Cinzberg's sense of the term. First, the aim ofthis chapter is to understand what is known about the occupational fateof gifted individuals. It is less concerned with ways to help students make careerchoices. Second, gifted children are typically identified by academic tests,and we need to consider carefUlly the relationship of academic ability toconventional occupational success. Finally, Americansociology has most commonly used occupational status and income to studythe questions that concern us. It is anestah-

142 134 Adulthood; Eatineare, Leadership, and Careers lished fact that not all occupations and not all human beings are regarded with equal worth in our society. For some reason, much of what is written about career education ignores these facts. This discussion does not.

Reasonable but Unrealistic Occupational Preferences The occupational preferences of schoolchildren have been studied for a long time. Surveys of adolescent occupational preferences have repeatedly yielded disqu.eting results. Unselected junior and senior high school students report an overwhelming preference for the professions (Roe. 1956). This phenomenon disturbs educators because thedisparity between job preference and job availability is so great. At a time when 6 percent of the population was employed inprofessional work, 50 percent of all students elected the professions as a possible occupational choice. Now elected manufacturing as a possible choice, although 50 percent ofall workers were at the time employed in manufacturing (Roe, 1956. p. 256). Similar results have been reported before and since Roe's research. These findings are disturbing because they suggest that the vast majority of young adults is unable to enact avowed preferences. (kcupational dissatisfaction threatens the stability of both fainily life and political life. One spends at least one-third of one's time working; dissatisfactionwith work is likely to affect the quality of relationships at home. Moreover, if the job dissatisfaction is shared among co-workers, it leads to labor un- rest. Clearly, the satisfaction that we derivefrom work is important in many ways. The finding is all the more distressing to educators because, according to Mar land (1974), Americans still subscribe to the seven cardinal principles of secondary education developed by the National Education Asso- ciation in 1918. The goals represented by these principleswhich include worthy home membership, a productive vocation, good citizenship, and worthy use of leisure time--cannot be fulfilled if cithens remain dissatisfied and unfulfilled in their work. This problem has been dealt with in two ways.

Fantasy.First, adolescents' job preferences have been finind to be- conic increasingly more llistic as students near the end of their school experience (Griffin. Kaheberg. & Alexander. 1981; Roe, 1956). Re- searchers have often concluded, reasonably enough. that fantasy plays an important part in adolescent development and that normal adolescents adjust quickly to the demands of adult reality. This approach suggests that the problem is not so threatening as might be imagined. This finding, however, merely reduces the magnitude of the problem; the approximation of job preference to job availability reimiins lop- 143 Adulthood: Eseixessee, Leadership, andCarrm 135 sided. Moreover, after age 18, job preferences change little (Roe, 1956), so that remaining dissatisfaction is likely to be a persistent problem.These facts suggest that it is very important to change students' attitudesabout jobs while they are still in school.

Career education.Therefore, a second approach has been to provide schoolchildren with information about the variety and natureof available jobs. One intention of this effort is to increase students' awarenessof the employment options that are potentially available to all citizens.Another intention is to apprise students of their abilities, both cognitive and noncognitive, and the requirements of various occupations (e.g., Ginzberg,1971; Marland, 1974; Winefbrdner, 1988). Educators hope suchknowledge will help individual students choose occupations that are actually(rather than potentially) available to them. These types of programs are carried on under the rubric of "career education." The term is misleading because it implies thatprofessional occupations (i.e.. careers) are potentially available to all students, amiscon- ception that it seeks to dispell. In fact, the types of conventional successin which virtually all Americans are most interested (high pay and high status) cannot be supplied to all students. Theconventional symbols of success are unfortunately in comparatively short supply. There are not enoughsuch jobs for everyone, and everyone wants them. Occupational satisfaction increases dramatically as security, pay. and status increase (Roe, 1956; cf. Terman &Oden, 1959). It is unlikely that the instrinsic worth of relatively tedious work can besubstituted for the extrinsic rewards of security, high status, and high pay. Someintellectually talented individuals may find important qualities in jobs that seemtedious to others, but these individuals willbe the exception. In general, satisfac- fion is a ctmcomitant of security, high status, and high pay.!n the current economy, however, the most tediousand lowest-paid jobs are experiencing the most rapid growth (Rumberger, 1984; Thurow,1987).

Professional status-seeking and gifted children.More than unselected students, gifted students prefer the professions. Tidwell (198U)reported that 91 percent of her sample of 1,593 gifted studentsexpressed such a preference; virtually all planned to attend college, with halfplanning to acquire master's degrees and one-fourth planning to pursuedoctorates. This combination of preferences and plans, when enacted,strongly conditions occupational status, regardlrs3 of IQ (Jencks et al., 1979).That is, students of average ability who carry out the same plansachieve oc tipational opportunities equal to those of their gifted peers. 'thegifted are more likely to carry out such plans than unselected students, but theirprofessional prospects are not so auspicious as onemight expect. For example, even in

144 136 Adubhood: Ensinexee, Leadership, and Careers Terman's sample, with a mean IQ of 151, manystudents did not become professionals (Terman & Oden, 1959). This fact is important because it reflects the stronginfluence of sonic variable other thP.n educational attainment.After the student's own level of educational attainment, his or her father'seducational level is the sociological variable that most strongly influencesadult attainment (Blau & Duncan, 1967; Jencks et al., 1979). Though mostgifted children do come from middle-class families in which parents arehighly educated, so do many other children. Children of above averageability (IQ 110-120) with advantaged family backgrounds doubtlessoutnumber children identified as exceptionally able, since 20 percent of allchildren fall in this IQ range. The competition for professional positions,however, is not even restricted to such above average students. Mostcolleges are not very selective; many undistinguishedstudents complete undergraduate andgraduate programs. The potentialfor high occupational status among thegifted is further compromised by the fact thatprofessional jobs are likely to become more scarce as thecybernetic revolution invades the professions(Rum- berger, 1984; Leontief, 1982; Wiener, 1954). The preceding discussion suggests threeconclusions:

1. The occupati;mai preferences of gilled studentsdiffrr in degree rather than in kind from those of unselected students. 2.Most gifted students have "realistic" occupationaland educational goals. 3. Professional jobs are scarce, and competition for them isintense.

Giftedness and Professional Status The previous discussion also suggests twopossibilities that need to be explored further. First, it is possible that mostprofessionals are not gifted individuals. Second, it is possible that many giftedsti.dents will not become professionals. The discussion that followsconsiders, first, the IQs of professional groups, and second, the careersof the highly gifted "Ferman sample ("Ferman & Oden, 1959). Examination of these possibilities is as important to anunderstanding of political, economic, and educationaltheory as to an understanding of the characteristics of gifted students.Differential education for gifted students is often justified by the claim that the gifted arefnture leaders (Fer- man, 1925) and that oureconomic system naturally favors the abilities of the gifted (Hollingworth, 1926; Herrnstein,1973: Hoyt & Hebeler, 1974). The research presented below suggests theneed for more plausible justifi- cations (cf. Bull, 1985).

The Qs of professional groups.The 1Qs of various professional groups have beenreported by several researchers in the past(Honing- worth, 1926; Herrnstein, 1973; Roe, 1956;Thorndike & Hagen, 1959; Jensen, 1973). In general the mean IQs ofprofessional groups are found 145 Aduktwod, Eminence, ltudenhip, and Careers 137 to exceed the national average.The more exclusive the profession, the higher the mean IQ of the group seems to be. Thus,the three most coveted and exclusive professional groups, doctors, lawyers,and engineers (cf. Kat- chadourian & Bo li, 1985), are typically found to have meanIQs varying from one to two standard deviations above thenational mean. Of the researchers cited above, Roe (1956)provides the most thoughtful discussion. She notes that in examining the IQsof occupational groups, one needs to consider notonly the means, but also the standard deviations and ranges as well. Roe notes that the IQ rangesof all occupations overlap substantially and that the IQ ceilings for alloccupations are similar. It appears that gifted individuals not only followall occupations, but also are represented in them. (see Box 5-2).

BOX 5-2Intellectual Talent in Occupational Groups: Means, StandardDeviations, and Ranges

r 1 1 1 7 I 1 Lawyer Auditor Teacher i=1=3 Stenographer 1 FJIIIIit Bookkeeper Clerk, general Cashier Manager, retail store Stock clerk Machinist Sales clerk Sheet metal worker Mechanic Cabinetmaker Bartender Welder C"I=D Painter, general Truck driver F armor 4 Teamster

1 10 20 30 40 50 51)713 00 90 100110120130140150150 170 AGCT score

(floc,r Roe, A 11,610. the p+vc hology t upations ip 72) Nirw Voris Wiley Used hs, permission

The mean ability of lawyers is about 125, the standard deviation 10,and the range from 95 to 160. The means for farmers and truck drivers, on the hand, were much lower (about 95), but the standard deviations were about 25, andthe ranges were about twice as large as for lawyers. The ceilings for the twolowest-status occupational groups were about 150. The distributionsfor both farmers and truck drivers were positively skewed, an indication that many subjects were marginally literate,

1 4 138 Adulthood; Eminence, Leadership, and Careers

Roe's observations address the two possibilities raised at the beginning of this section. Although a large percentage of professionals score in the gifted range of general intellectual ability, most do not; moreover, many highly gifted students will not become professionals. Since relatively few persons are employed in the professions and since the vast majority of' persons is employed in nonprofessional work, it is possible that more gifted individuals are employed in nonprofessional than in professional jobs. This possibility may be more likely if one includes women, who are still underrepresented in the most coveted professions (cf. Fleming. 1985; Rum- berger, 1984). These conclusions are consistent with the findings of Jencks et al. (1979). Being gifted is not sufficient or necessary for achieving entry into any profession. Even being gifted and well-educated is not sufficient to ensure professional status. Competition for professional jobs is intense, and factors other than academic ability may be of strong interest to employers.

Careers among Terman's subjects.Lewis Terman and Melita Oden published the 35-year follow-up of ,.heir very high 1(2 sample in 1959. At the time data were collected (1949-1954) the subjects were about 44 yeio-s old on average, and their occupational patterns were therefore considered to be well-established (Terman & Oden, 1959). Because most of the data considered in this section of the chapter are based on the occupations of men, discussion of women among Terman's sample is given less attention here. At the time of the study, women suffered great sex discrimination in employment, a condition that has improved only somewhat since 1959. Interested readers are urged to consult Sears and Barbee (1977) for a discussion addressed particularly to the Lite of women in the Terman sample. In interpreting the following discussion, readers should recall that Terman and Oden's subjects were a highly gifted cohort. Terman and Oden (1959. p. 74) report that 45.6 percent of their sample were employed in the professional category of the Minnesota Occu- pational Scale. This figure excluded business executives and others whose activities would now (because of the rise of postbaccalaureak business schools) be construed as prokssional. 'these figures can nonetheless be compared to Roe's (1956) data cited previously, and they suggest that the gifted are far more likely to achieve prokssional status than unselected persons. Since most of the gifted sample completed college (70 percent of the men and 67 percent of the women), a more germane comparison involves plied college graduates versusunselectedcollege graduates. In order to Coln- pare the professional employment of their sample to the professional employment of unselected college graduates, 'Ferman and Oden combined three categories from the Minnesota occupational Scale. The resulting

147 kiss:Mond: Emthsessre, Leaders*, and Careers 139

"professional" group, defined more loosely than onthe Minnesota Occupa- tional Scale, accounted for 94 percent ofall gifted college graduates in the sample. The comparable figure forunselected U.S. college graduates was 84 percent, according to 'Ferman and Oden(1959). The occupational advantage enjoyed by Terman'ssubjects was slight in comparison to all college graduates. The occupational attainment ofall gifted men in the Terman sample about equaled the occupational attainmentof both male and female college graduates, according to Termanand Oden's figures (1959, p. 80). Terman and Oden's data make it apparentthat, depending on the definition of "professional," many subjectsdid not become professionals. Sometimes these were students who didpoorly in school or did not continue their education beyond high school.Sometimes they were college graduates who accepted lower-status jobs as a matterof choice (Ferman & Oden, 1959). Some observers have noted that mostjobs, even professional jobs, may require moreeducation and higher levels of skill than arecalled for by the demands of the job (Berg, 1972; Bowles& Gintis, 1976; Jencks et al.. 1979). This observation may be particularly truein the case of gifted students; fbr them, even the most covetedoccupations may prove somewhat limiting or unfulfilling (Gold, 1965; Willings,1985). About half of Terman and Oden's (1959) male subjects reported"deep satisfaction" in their work; about 40 percent were "fairly contented";the rest were not particularly satisfied or were discontented. Doctors,bankers, and colic cp teachers were more satisfied than the averagegifted subject; authors, enginc:rs, clerks, and skilled tradesmen were less satisfiedthan average. Levels of satisfaction were lowest among authors (20 percent),skilled tradeworkers (18 percent), clerks (17 percent), andschool teachers (10 percent). Among 'Ferman and Oden's sample. at alloccupational levels, there appears a type of-individual whose behavior attests to the existence of compelling intellectual interests not fulfilledby occupational activity. These individuals tended to find more meaningin their "avocational" pursuits than in their occupations. Anecdotesabout these individualshouse- wives, executives, teachers, lumsepainters. carpentersreportthat in an attempt to accommodate theirintellectual interests, they often made chokes that could be judged economically unwise orimpractical. The significance of this characteristic' will be explored furtherin our discussion of intellectualism among college students. It is not possible to conclude from 'Fermanand Oden's (1959) data that giftedness is a substantial occupationaladvantage. 'Ferman and Oden believed it was, of course, but more recent work (e.g.,Baird, )985; Jencks et al., 1979) documents only aslight positive overall association of IQ with occupational performance, once the effect ofeducational attainment is controlled. Jencks et al. (1979) found that althoughthere is a 0.47 correlation between IQ and occupational status, 60 percent to80 percent of this

148 140 Adulthood: Emiarnee, Leadership, mod Careers correlation is attributable to the amountof schooling. When anumnt of' schooling is controlled, variations inintellectual ability account for only about 8 percent of the variancein occupational status. This findingis consistent with the reinterpretationof Terman and Oden's (1959)data.

Conclusions and cautions.Though many gifted students become professionals, perhaps an equal number do not;though many doctors and bankers are exceptionally talentedintellectually, many are not. A one to one correspondencebetween exceptional intellectualability and professional status does not exist. The l(2s ofprofessional groups and the professional status of highly giftedstudents illustrate the influenceof factors other than academic ability onoccupational status. Intellectualtalent. though important, is only one kindof talent needed to achieve entryinto a profession. Talents unrelated tointellectual ability are probably ofequal or greater concern toprospective employers or colleagues(Baird, 1985; Jencks et al.. 1979). The free enterprise system determinesthe nature and distribution of jobs for aspiring students, and that factis often overlooked by clreer education. Schools may function tolegitimate the distribution of scarceand desirable jths to advantaged students(Meyer. 1977); moreover, suet. a function seems to take surprisingly little noteof exceptional talent (Howley et al., 1986). Nonetheless, toconclude that schools should cultivate noncog- Mtive characteristics that would help morestudents acquire well-paid jobs of high status would be unwise. As thefollowing discussion indicates, many gifted students possess intellectual abilitiesand interests that cause them to regard employmem prospects skeptically(Katchadourian & Boll. 1985).

Intellectualism and Careerism The job market for college graduatesis considerably worse than it was in the 1960s (Rumberger, 1984).Professional positions are now more difficult to secure, and many collegegraduates take nonprofessional jobs. In a survey of job prospectsfor college graduates in the years1960-1990. Rum- berger (1984) notes that the employment prospectoi. college graduates no longer resembles that of college graduatesdurMg the 1960s taut more closely approximates the prospects ofcollege graduates during the 1950s. This survey confirms the impressionof many observers. Many observers are convinced that morecollege students than ever are anxious toacquire lucrative jobs after graduation. Theseobservers fear that students often hold such careerinterests to the exclusion of cultivating intellectual interests and values. Thispossibility concerns liberal arts colleges particularly. Katchadourian andBoll (1985) studied careerism and intellectualism among a cohort of StanfordUniversity undergraduates. Their findings about the characteristicsof the Stanford cohort should be of considerable interest to teachers of the gifted.

1 4 Adulthood: Eusisseser, Leadership, and Careers 141

The Stanford study.Katchadourian and Bob (1985) were interested in discovering the extent to which careerism and intellectualismmotivated their undergraduate students. From 1976 to 1983 they followed a 20 percent random sample of students (n = 320)in a single class (the class of 1981), collecting data on career choices, academic majors, grades,sociological background. and so forth. They also conducted structuredinterviews that are quoted extensively in their report.Careerism and Intellectualism Among College Students(see Box 5-3).

BOX 5-3Careerism and Intellectualism at Stanford

Most students in the Stanford sample (72 percent) came from public highschools. Their mean Scholastic Aptitude Test (SAT) scores wereSAT-Verbal, 600, and SAT-Mathema- tics, 654. (In 1976 the SAT national means were 429 and 470,respectively.) The parents of most students were well educated; family income washigh compared to the national median (18 percent came from homes with an annual income above $100,000; 24 percent from homes with an annual income below$30,000.) The high cost of attending Stanford doubtless explains this skew.Though it is not a sample of gifted students, the gifted seem to be as well represented at Stanford as they are in the most coveted professions. Some students scored high on both scales and some students scored low on both scales. As a result the researchers developed a four-part typology: intellectualsscored high on intellectualism and low on careerism; careerists scored low on intellectualism and high on careerism; strivers scored high on both scales;unconnected students sco-ed low on both scales (see Figure 5-1).

FIGURE 5.1Typology of College Students (from Kafchadourian and Boll, 1985, p. 34)

Careerism

Intellectuals Strivers

Intellectualism

Unconnected Careerists

Using factor analysis, Katchadourian and Bo li (1985) developed unrelated (r = 0.10) scales for intellectualism and careerism. The scalesused items that addressed three basic concerns: reasons forattendiag characteristics desired in a major, and characteristics desired in a career. For example, students who scored high in careerismcited "acquisition of marketable skills," "future financial security," and "establishingprofession- al contacts for the future" as the prime reasonsfor attending college. Students who scored high in intellectualism cited "learning tothink nit-

15o 142 Adulthood; Essioessee, Leadership, and Carters ically," "'pursuing] general liberaleducation," and "developing artisticand esthetic taste and judgment" as prime reasonsfor attending college (Kat- chadourian & Bob, 1985. p. 31). Careerists tended to be "mercenary"in their pursuit of a coveted career (Katchadourian &Boll, 1985, p. 102). Earning a lot of moneyquick- ly was a goal for many. Most careeristsregarded their studies as unpleasant work necessary for achievingprofessional goals. They often expressed boredom with academic work, whileacknowledging Stanford's usefulness as an "elite finishingschool" (Katchadourian & Bob, 1985, p.86). They were largelyuninterested in social and political issues.Overall, careerists tended to have the highest educationalaspirations of all four groups: 50 percent of each sexplanned on earning doctorates (in spiteof the fact that they regarded academic work asdistasteful). Intellectuals, on the other hand, regardedlearning as an activity pleasurable in itself. Intellectuals were veryinterested in social issues, but tended to be political moderates ratherthan progressives. Their occupational ambitions were directed at securing workthat was interesting rather than lucrative. They were much more likelythan careerists to change plans after starting college. These changes resultedin movement into humanistic disciplines. whereas the net changes ofother groups drew students away from humanistic disciplines. Although strivers espoused strongintellectual interests, their educational aspirations, career choices,and college majors resembled those of' the careerists. In general, their strong careerinterests seemed to take pre- cedence over intellectual involvement. When the researchers examined the data onunconnected students, however, they identified a subgroup whoresembled the intellectuals. Mem- bers of this subgroup were interested inand active in pursuit of intellectual issues, but they failed to endorse Stanford'sinstitutional perspective. Kat- chadourian and Bob (1985. p. 192) calledthese students the "academic dissenters": the politics of many could hedescribed as progressive rather than moderate. Of all groups, theunconnected were the least likely to choose business as a career: afterthe intellectuals. the unconnected were the group most likely to choose humanisticprofrssions (writing and jour- nalism) and teaching.

laspikalions.The Stanford study documents deardifferences in values among a group of very talentedundergraduates. The findings are consistent with findings about the careerchoices of gifted high school students (Marshall, 1985) and aboutgifted adults (Willings, 1985). Kat- chadourian and Bo li's findings provide furtherevidence that the pursuit of theoretical and aesthetic interestsmotivates many gifted individuals. regardless of employment prospects.While it is probable that many gifted

151 Adulthood: Emanates, Leadership, and Careen 143 students will resemble the Stanfbrd intellectualsand academic dissenters, it is also certain that many others willresemble the careerists. The careerists, however, seem least in need of careereducation, whereas the intellectuals are intent on makingtheir own way. Gifted Students, Intellect, and Occupational Life As they mature, students withexceptional academic and intellectual talents begin to function as adults. Someof these students are fbrtunate enough to enact comparativelyindependent decisions about the useof their talents. For others, parentalvalues and social conventions overtly determine their actions. The difference in occupationaloutlook is perhaps inherent in the nature of intellectual talent,which may either develop itself or beapplied to a practical task. Themotivation for self-development is intrinsicinterest, which seeks understanding rather thanacquisition. By contrast, the motive of practical intelligence seems to beacquisition. "The dominant value orientation anmng hard-core careerists is tomake money: a lot of it, and fast" (Katchadourian & Bo li, 1985, p. 102). Itis probable that the gifted high- school careerists observed by Marshall(1985) shared this motive. The demands of occupational lifeobviously favor the application of intelligence to practical tasks. All adults must cometo terms with this fact. For frost students, whose intellectualvalues are less extreme than among the gifted, the reckoning is not sodifficult. The findings cited above suggest that gifted studentsresolve the occupational dilemmain two fundamental ways. Intellectualism and vocation.l'he difference between intellectuals and careerists is perhaps best illustratedby the term "vocation." A vocation is literally a "calling," and intellectuals seem tobe engaged more in the process of herding a call,that is, of discovering a vocation, than inacquiring a career. The Stanford intellectuals and academicdissenters constitute a group whose primary intellectual activity is to constructmeaning. Two characteristics of this group are of' particularrelevance. First, students of this type have intellectual interests that spanseveral fields. 'Fhey do not become specialists easily or happily (cf. Wi flings.1985). Second, this type of student typically possesses strong verbal skills.Exercise of those skills may be what draws such students to the humanities.They are less interested than other students in math or science. Thesestudents hope that their intellectual interests will eventually lead them todiscover occupations that are intrinsically rewarding. The conventional successrepresented by lots of money seems to be much lessimportant to this type of student than to the careerists.

152 144 Adulthood Essimmee, Leadership, msd Careers

Careerism and anti-intellectualism.Treating the intellect solely as an instrument that is useful in securing entry to lucrative professionsseems to entail denial of the intellect. Careerists do not even expectto enjoy their studies. Many careerists apparently ignore theirown perceived need for intellectual self-development (cf. Katchadourian & Bo li, 1985,p. 107). The Stanford researchers were most distressed with these students' failureto discover wider intellectual interests. The most striking suggestion, however, in Katchadourian and Bo li's account of the careerists is that their anti-intellectualism may be imposed from without. Careerists are strongly compelled by extrinsic influences. Parents often exert authority overtly by such manipulationsas threatening to withhold financial support (Katchadourian & Boll, 1985, p. 96). Stu- dents are thus conditioned by their family values to make practicalcareer decisions early and to stick with those decisions. Coleman (1961,p. 304) observed that a similar process pulled even average high-school students away from scholarship.

Occupational prospects far the gifted. A talented studentcan choose to ignore the call of the intellect for self-development and opt single-mindedly for the best-paid occupation. Alternatively, a talented studentcan ignore the issues of practicality and pursue intellectual interests single- mindedly. It seems, however, that whatever route gifted students choose, they will inevitably feel misgivings. Some careerists regret that they darenot follow more compelling studies (Katchadourian & Boli, 1985);some intellectuals regret that their choices lead to less well-paid occupations. Many gifted students will share in both regrets. Teachers ought to recognize that only a fortunate few will find a comfortable existence doing well-paid, meaningful work. Teachers of gifted students should understand that the characteristic of exceptional intellectual ability does havesome practical value in the occupational world. They should also realize that the machinations of the economic system overwhelm the practical value of the intellect. It isnot clear, for example, that exceptional ability is a greater practical occupational asset than moderate ability. A fortuitous combination of moderate ability, the right background, and occupationally relevant noncognitive characteristics is probably a more effective influence on occupationalstatus than extren.2 ability alone. It is questionable whether reliable diagnosis of such fortuitouscom- binations of characteristics will ever be possible. Considering the scarcity of desirable jobs, the changing nature of the job market, and the continuing restructuring of the economy, it is also questionable whether public elementary schools and high schools should be the institutions to identify and groom students for occupational success. 153 Adsdthead; Essismnee, Leadership, assd Careers 145

SUMMARY

This chapter examined the adulthoods of children identified as gifted by the schools. It began by developing the notion of a continuum of talent in which substantial adult success (leadership, eminence, or career success) was viewed as a phenomenon more rare than childhood giftedness. The relationships between giftedness and these three varieties of adult success were explored in depth. In general, the chapter noted that intelligence does not predict adult success very well. It examined the historical and cultural influences that may account for the weakness of the observed relationship. The discussion of eminence and giftedness distinguished among fame, renown, and eminence. The relationship between eminence and intellectual traits was viewed critically through the work of C. Wright Mills and Richard liofstadter. The discussion concluded with an historical review of research on eminence, contrasting the work of Galton with that of later researchers. The chapter next examined the relationship between leadership and giftedness. The discussion reviewed significant historical models of leadership (in the works of Plato, Machiavelli, and Locke) and major contemporary leadership roles (corp)rate, political, and cultural). The importance of the typology of transactional and transformational leadership to gifted education was examined critically. The chapter concluded with an investigation of careers among the gifted. Like average students, gifted high-school students express preference for professional careers. Nonetheless, studies indicate many gifted students do not become professionals. Two types of career interest were distinguished in the discussion: (1) strong professional commitment based on the desire to become affluent and (2) other career interests based more on a search for satisfying work.

154 Academically Talented Children

I.Focusing Questions II.Relationship Between IQ and Academic Talent A.High-IQ Children 1.Academic achievement of high-1Q children 2.Aptitudes of high-IQ children B.Academically Talented Children 1. Verbal aptitude 2.Mathematical aptitude 3.Science aptitude 4.IQ scores of academically talented children 5.Weaknesses in correlational studies 6.Sources of individual differences C.Special Talents 1. Talent in prerequisite academic skills 2.Talent in traits thought to describe cognitive processes II.Verbal Precocity A.School-related Characteristics of Verbally Precocious Children B. Receptive Language Skills and Giftedness 1. Gifted students' reading achievement 2.Interest in reading: then and now C.Speech and Written Composition of Gifted Students 1.Oral language of gifted students 2.Written composition 155 146 Academically Takosted Children 147

3. Penmanship as a confounding variable ir composition D.Foreign Language Aptitude 1. Verbal intelligence and foreign language achievement 2. Other correlates of foreign language achievement E. School Responses to Verbal Precocity 1. Identifying verbal precocity 2. Placement 3. Program characteristics 4.Acceleration, enrichment, and instructional level 5. Programming cautions 6. Trends for the future III.Mathematical Ability A.Mathematical Skills 1. Calculation 2. Concepts 3. Logic 4.Problem solving B. The Characteristics of Mathematical Ability 1. Krutetskii's study 2.Mathematical ability according to Krutetskii 3. Quantitative traits not part of mathematicalability 4. Gender and mathematical ability C. School Responses to Exceptional MathematicalAbility 1. Identification 2. Placement 3. Enrichment 4.Acceleration 5. The dilemma of school responses to mathematical ability IV.Summary

Focusing Questions 1. How is verbal precocity related to intellectualability as measured by an IQ test? 2. How do language comprehension andlanguage production differ, and how does each relate to IQ? 3. How should schools respond to the needs ofverbally precocious students? 4. What are the characteristics of mathematicallygifted students? 5. How do verbal ability and spatial ability relate tomathematics achievement? 6. How should schools respond to the needs ofmathematically gifted students?

1 56 148 Academically Talented Children

RELATIONSHIP BETWEEN IQ AND ACADEMIC TALENT

The moderately high, positive correlation between IQ and academic achievement is well established. The average correlation varies from about 0.50 to 0.80 depending on the measurements used (Vernon, 1979). Chil- dren's standardized achievement test scores usually correlate more highly with their IQ scores than their grades do. The size of the correlation also depends on the range of scores. When a wide range of IQ scores and academic achievement levels is considered. IQ scores are "remarkably accurate" in predicting academic achievement (Messe, Crane, Messe, & Rice, 1979). When the range of IQ or achievement test scores is restricted to a small portion of the distribution of test

BOX 6-1Schematic Representation of IQ and Academic Abilities

Exceptional mathematical ability

1 1Exceptionat t general ) academic ability /(high 10)

1 Exceptional verbal ability

DESCRIPTION

A, B, and C are distinct sets of students: the three sets do not intersect.

A + B + C = all students with significant academic abilities A + B = all students with exceptional mathematical ability B + C = all students with exceptional verbal ability A z= students with exceptional mathematical ability only B = students with both exceptional mathematical and exceptional verbal abitity C students with exceptional verbal ability only

1 5 7 Aeadnakaliy Talented Children 149 scores, as it is instudying gifted children, the correlation islower. The purpose of this sectionof the chapter is to discuss likely reasonsfor these effects and to consider the relationshipbetween high IQ and talent in specific academic areas. In this section and throughoutthe book, the terms, "talent," "aptitude" and "achievement" are used nearlysynonymously. Talent may be exhibited as academic achievement or asacademic aptitude. When referring to the concept of genetically endowedability, we use the phrase "innate ability." See Box 6-1for a schematic representation.

High-IQ Children High-IQ children are, by definition, academically aptin general (see Chapter 2). Their general aptitude stronglyinfluences, but does not determine, their performance in the classroomand on standardized achie, e- ment tests. They usuallydo well in school, but some high-IQ childrendrop out of school, earn poorgrades, achieve far below expectancy, orexhibit mediocre mathematical aptitude.

Academic achievement of high-Ie children.Although high-IQ children usually earn above-average achievement scoresin most acadennc areas, their achievement test scores are not asfar above average as their IQ scores, As a group, they scorehighest in reading comprehension and language usage. They are alsosuperior in spelling and mathematics, but their achievement in these areas is sometimessignificantly lower than their IQ scores. High IQ scoresshow little or no correlation with achievementin school subjects that emphasize motor skills,and gifted children often make only average grades in penmanship orphysical edtwation, for example. This achievement pattern was first reportedby Terman (1925). whose gifted subjects' achievement on astandardized test was highest in language usage, next highest in reading, thenspelling, and lowest, though still far atxwe average, in arithmetic. Thesubjects' mean language quotient was 147, whereas theirarithmetic quotient was 137. Their mean IQof 151 was 25 percent higher thantheir mean academic achievement score. Most studies show similar results: Gifted children'Aachievement score are typically tem extreme than their /(2 scorei. Gallaglwr and Crowder's (1957) study ofhigh-IQ children in the regular classroom found most of themachieving above grade level. Like Terman's subjects, these children's scoresin reading comprehension were highest; their scores in arithmetical computationlowest. Sixty-eight percent of this group of children,who bad a median IQ of 150, scored more than three grade levels above their chronological agegrade placenwnt on paragraph meaning. Fifty-eight percent scoredthree or more grade levels above on science; but only 27 percent scoredthree or num- grade levels

158 150 Academically Takata Children

TABLE 6-1Whie-Ranse Achievement Text Scores of Gifted Children in Grades 4, 5, and 6

READING ARITHMETIC SPELLING

Grade 4 8.2 5.8 7.1 Grade 5 8.7 6.8 7.6 Grade 6 9.9 7.2 8.5

Source: Adapted from Wharton, Karnes, & Currie, 1985, p. 151.

above in arithmetical reasoning. Very few, only 3 percent scored threeor more grade levels above grade placement on arithmetic computation. A study by Hollingworth and Cobb (1928) suggestsa possible explanation for gifted students' relatively low scores on computation. There- searchers speculate that the more complex the task, the grrater the advantage uf high IQ. In simple tasks, such as addition of whole numbers, for example, high-IQ students seem to enjoy little or no advantage in comparisonto average students (Hollingworth & Cobb, 1928; cf. Krutetskii, 1976). The achievement pattern of gifted children with IQscores lower than 150 has also been established (Whorton. Karnes, & Currie, 1985). Agroup of gifted elementary-56mA children with a mean WISC-R IQscore of 131 was found to perform two to four grade levels above actual grade placement on differem subtests of an achievement battery. The children scored as shown in Tab.& 6-1 on the Wide Range Achievement Test. All of these scores are above grade level, but nearly all are below the achievement level.3 that might be expected on the basis of IQ. Arithmetic scores are much farther heiow the expected level than are reading and spelling scores. This pattern of' performance is similar to that of Terman's (1925) and Gallagher and Crowder's (1957) findings, despitea 20-point difference in the mean IQ scores of' the samples.

Aptitudes of high-IQ children.High-IQ children usually score above average on tests of special aptitudes, but their performance varies greatly depending on the test and the type of aptitude. On tests of verbal and quantitative aptitude, they score as high as students whoare several years older (Martinson, 1972). On tests of other kinds of aptitude, for example, artistic, musical, or mechanical, their scores are closer to thenorm. Wilson's (1933) study of* h;gh-1Q children's perfbrmanceon special aptitude tests found their mean scores to be above averageon every test administered. On tests of* art judgment, musical ability, science achievement, and mechanical aptitude, this group of children, whose mean IQ was 150, scored higher than the mean for children theirage or a yeut' or two older. 'flie children scored highest on a test of*muskmemory. They scored

15 z4 Academically Talented Children 151 lowest (but still above average) on a test ofMusical accomplishment, apparently because superior performance on that testdepended on rather extensive instruction in music (Wilson, 1953). Thedistribution of their scores on all of the tests, however, was closer to the normthan the distribution of their IQ scores; and the correlation between their aptitude test scoresand their IQ scores was, for the most part, low to moderate(ranging from 0.16 to 0.57). A study of science aptitude in six- andseven-year-old children with a mean IQ of 150 also foundlow correlations between IQ score and special aptitude (Lesser, Davis, ar Nahemow, 1962). Thechildren's Stanford-Binet IQ scores correlated 0.21 with a battery of sevencriterion-referenced science achievement tests. ltaveraged about 0.10 with two forms of an experimental science aptitude test. The aptitude test wasreported to be a better predictor of science achievement than the IQ test. Therestricted range of' the group's IQ-test scores may, however, accountfor the low correlation between IQ and science aptitude and science achievementin this study. On the other band, the similarity between the contentof the science aptitude and achievement tests (both were based onthe school's third grade science curriculum) probably accounts for the high correlationbetween those two measures. The studies discussed above suggest that, amonggifted children, IQ is not a good predictor of' narrowly defined aptitudes.This conclusion suggests that children with IQsof' 130 are as likely to have high aptitude scores as children with 1Qsof 100. Similar findings, you may recall, characterize the relationship between IQ and divergentthinking.

Academically Talented Children Academic achievement is usually categorizedaccording to school subjects, whereas academic aptitude is typicallyconstrued nuire broadly. Tlw two major varieties of ;teat;emir aptitude are mathematical (or quantitative) and verbal (see Box 6-1). Occasionally, scienceaptitude is differentiated from mathematical and verbal aptitude. Both mathematical and verbal skills areprerequisites to success in many academic lidos,including science. Science aptitude is investigated primarily because it is perceived to be in the nationalinterest, rather than because it reflects distinct cognitive skills. Successin science courses, however, like success inhistory courses, can largely be accounted for by quantitative and verbal abilities (Brandwein, 1955;Howley et al., 1986). Although scores on verbal and mathematics tests arepositively related, there are many talented students with alarge discrepancy between their verbal and mathematical aptitude. Someverbally gifted students have only average mathematical aptitude; somemathematically gifted students 152 Academically Talented Children have only above average verbal aptitude. The more extreme a student's verbal aptitude, the less likely it is that the student's mathematical aptitude will be equally extreme, and vice versa. Nonetheless, exceptional aptitude in one category (general. verbal, or mathematical ability) usuallyindicates above-average aptitude in the others (see Box 6-1). Verbal aptitude is usually associated with advanced comprehension and facility in language usage. Verbally apt students are likely toperform better on 1Q tests than on tests of mathematical aptitude, since IQ tests sample verbal material more heavily than quantitative material (Oiler & Kyle, 1978).

Verbal aptitude.Several distinctions are relevant to the examination of verbal aptitude. Language comprehension and language production implicate different levels of performance and different sorts of tasks. Learning foreign languages requires good skills of perceptionand auditory memory not associated with advancedcomprehension or with most intelligence. constructs.

Mathematical aptitude.Tests of mathematical aptitude correlate modtTately with g. Spearman's general intellectual factor, probablybe- cause they measure conceptual knowledgeand reasoning skills. Excellent (as opposed to merely adequate) computation skills are not,however, related to high levels of mathematical ability (Hollingworth & Cobb, 1928; Gallagher & Crowder, 1957; Krutetskii, 1976). High-1Q students' mathematical achievement may also be less extreme than their 1Qs andtheir reading achievement because such students may depend on classroom instruction to learn arithmetic. There is no mathematical equivalent ofread- ing difficult works independently, the experience that is probablyresponsi- ble for the exceptional reading achievement characteristic of so many gifted students.

Science aptitude.Different scienceshiology, chemistry. physics, psychology, sociologymake very different demands on the quantitative and verbal abilities of students. Whether or not a single construct("science ability") underlies achievement in subjects so diverse is not clear. Though mechanical and spatial abilities (Roe, 1953) and analytic abilities (Guilford. 1950) are presumed to be important to achievement in some naturalsci- ence courses, their predictivevalidity has not been demonstrated. It is not even clear that chemistry andphysics, for example. make similar demands on these skills. Investigatitms of science aptitude concern adult accomplishment more than school achievement (Holland &Astin, 1962; Holland & Richards. 1965; Taylor & Ellison, 1975). This focus is a direct result of the need to supply accomplished scientists in the national intert. In all cases 161 Academically Talented Cltildren 153 the individuals who are the subjects of these studies are all 'er' talented; thus, a very high levd of academic achievemem is a prerequiAite to a high level of adult sdentific accomplishment (Baird, 1985).

IQ scores of academically talented children.Few studies have investigated the IQ scores of children identified as gifted on the basis of exceptional verbal or mathematical aptitude; however, these few studies have consistently found academically talented children to score at least above average on intelligence tests and other testsof cognitive development. Stanley (19761>) reports that none of the children identified as mathematically talented by their high scores on the Scholastic Aptitude T sI- Mathematics is of average or below-average intelligence. In a case studs' of* seven junior-high students identified asacademically talented by their performance in a science competititm, Fox (197tia) found that the students scored far above average on two measures of intelligence, Terman's Con- cept Mastery Test (CMT) and the Raven's Progressive Matrices.keming ( I 97(k) assessed the performance of mathematically talented children on a Piagetian test of cognitive development and found that the talented lift h- grade students performed at a more advanced level than average seventh- grade students. Though Piagetian measures do not yield intelligence quotients, superior performance nm Piagetian tasks is nsually assot iawd with high IQ scores (see (:hapter 8).

Weaknesses in correlational studies.There are several factors that must be taken into account when interpretingcorrelational studies of gifted students: ( 1) many studies of gifted children Yield low correlations because the ceilings of the tests (IQ. aptitude. or achievement) are um low to represent accurately the gifted children's performancelevel; (2) even when test ceilings are sufficiently high, extreme test scores are associated with the greatest amount of' error in tneasurenwnt; and (3) the construct validity of*some IQ, aptitude, and achievement tests is (inestimable. To an unknownextent all of these limitations confound the measured relationship between intelligence and academic talent.

Sources of individual differences.Among chikiren who have comparable IQ scores, differences in performance on measures of at ademk aptitude or achievement may stem from any of the three possibilities mentioned above; or they may reflect real perfOrmance differences caused by ( I ) differences in instruction. (2) differences in interest, or (3)differences in innate ability. Instruction in school has a much greater effect on aptitude and achievement test scores than on IQ scores. I figh-IQ children who have been provided superior or additional instruction in mathematics. for example, acquire more advanced knowledge and so score higher on measures of

162 154 Accuirmtkaily Talented Chi idnew mathematical aptitude or achievement (Keating & Stanley. 1972). By contrast, poor instruction or limited access to advanced instruction reduces the likelihood that gifted children's math performance will correlate highly with their IQ scores. Although it is difficult to determine the direction of the causal relationship, it is apparent that children's interests and their achievement are related. Most likely, interest and achievement are interactive, with higher achievement promoting higher interest, and higher interest, in turn, effecting higher achievement. Even as children, internationally acclaimed mathematicians, musicians, and athletes spend much of their time engaged in the kind of activities in which they later excelled (Bloom, 1982; 1985). Adult scientists often report childhood interest in mechanical and scientific projects (11:-,e, 1953). Both scientifically and mathematically talented children express greater interest in theoretical ideas than average children (Haier & Denham, 1976; Krutetskii, 1976). Verbally gifted students are also highly interested in theoretical ideas, but they express more interest in aesthetic ideas than do mathematically or scientifically talented children (Fox, 197(k). The existence of innate abilities has not yet &en denumstrated con- vincingly. Instead, their existence has been inferred, principally fmtn Ob- served differences in the performance of schoolchildren on standardized tests. The inferences of numerous studies reinforce popular opinion;and many educators and lay persons attribute all sorts of individual differences to heredit} (Gould, 1981). The work of Benjamin Bloom (1964.1971, 1976, 1982. 1985), however, illustrates an alternative. His recent work, a study of 30 highly accomplished musicians, athletes, and mathematicians. suggests that upbringing most strongly influences the development of tal- em. (For a detailed discussion of- nature and nurture. see Chapter 2.)

Special Talents Some special talents seem to bear little relationship to the major varieties of academic talent discussed previously. These special talents seem to have limited value in predicting significant academic accomplishment. They relate to (1) prerequisite academic skills and (2) cognitive or perceptual processes.

Taknt in prerequisite academic skills.Students who are exceptional spellers and "lightning" calculators have served to entertain professional and lay audiences for centuries. The study of these special talents is pertinent to basic research in psychology, but is, however, largely unrelated to the cultivation of academic talent in schools. Excellence in some school subjects, such as spelling or arithmetical calculation, does not predict academic talent because a moderate level of success (i.e., competence or mas- Academically Tcdeared Children 155 tery) in these subjects is all that is required toachieve excellence in later academic work. These elementary skills make verylimited demands on students' intellectual abilities. Many high-IQ children exhibit only slightly above average or even average achievement in spelling,word attack, arithmetical computation, or penmanship. Gifted children have something of areputation for poor penmanship, in fact (Freeman, 1979). Some high-IQ children arebelow average or deficient in some ofthese skills. It is even possible fOr students to exhibit below-average achievementin all these skills and still score in the gifted range on IQ tests. These skills are hardly sampled atall on IQ tests. Thus, the prominence of prerequisite skills in theelementary curriculum may explain the difficulty ofidentifying gifted elementary-school children On the basis of' classroomperformance or observation. Tasks that require comprehension, reasoning, and interpretation seem to be moreclosely associated with the major varieties of academic talentthan skills that depend on motor coordinatim, visual perception, or rote memory.

Talent in traits thought to describe cognitive processes.Complex intellectual tasks obviously require the integration of anumber of skills. There is a problem, however, in defining separate abilitiesfor all such skills. A larger number of abilities necessarily results inthe diminished predictive validity of' each new ability (Krutetskii, 1976). Narrowcognitive traitssuch as spatial scanning, perceptual speed, spatialorientation, and flexibility of dosure (Carroll, 1976) or Guilford's (1959) 120different intellectual abilitiesare of very limited use in predictingoverall academic talent (Carizio & Mehrens, 1985; McNemar, 19641 Testingfor such skills is primarily of' interest for basic research. Often research findingsyield insights into puzzling encounters in the classroom, but a great many questions must be answered befOre educators can trust thesefindings to guide educational practice. The practical status of some broad, nonverbal, cognitiveskills (e.g., spatial and mechanical reasoning) is even more equivocal.Though spatial and mechanical reasoning skills seem relevant to a varietyof fields (such as engineering. architecture, astronomy, geology, and fine arts),their func- tim in predicting talent is questionable. Krutetskii (1976),fOr example, believes that spatial ability is not a necessary element of'exceptional mathematical ability. No academic disciplines appear to be soheavily dependent on these skills that their assessmentpredicts talent more effectively than IQ, verbal, or quantitative measures. Sometimes special abilities are commended as a vehicle withwhich to improve the fairness of identification efforts. It has beenhypothesized. for example. that disadvantaged or handicapped children showsuperiority on nonverbal tests of cognitive processes but not on academicaptitude tests. Nonverbal reasoning ability, however, predicts achievementin an academic

164 156 Academically Talented Children discipline less well than tests that sample skills that are more st;ongly represented in most academic disciplines. Children identified a..iaiented by their scores on nonverbal reasoning tests,rather than on tests of academic achievement and aptitude, must be provined extraordinaryinstruction to upgrade their verbal and mathematical skil!s.

VERBAL PRECOCITY

Verbal precocity, or exceptional verbal ability, helps ensurehigh scores on IQ tests. strong verbal-logical skills also characterizemathematical ability (see the folifiwiog r,..-ction). As we noted above,gifted children ()lien exhibit both exceptional verbal aptitude and exceptional mathematicalaptitude. In fact, so few studies have investigatedexceptional verbal ability as titshml from high IQ that itis difficult to discuss verbal precocity as a separate phenomenon. Our discussionof verbal precocity, therefore, fix uses primarily on the performance of high-IQchiklren in reading, writing, speaking. and foreign language learning. School-related Characteristics of Verbally Precocious Children Tests of verbal ability and aptitude, such as the verbalportion of the Scholastic Aptitude Tes; (SAT-V) and the Stanford-BinetIntelligence Scale. are good predictors of school success. Oneof the most significant st turd characteristicsof verbally precocious children is that they are usuallY successful students. Among middle-class students who scorein the 97th percentile or higher on the Stanford-Bilwt Intelhgence Scale, alxmt !JO percent can be expected to have agrade point average of 3M CB-) ot higher (Pegnato Sr Birch, 1959). School success is partly a functiim of these children'ssuperior comprehension of what thev read. Many verbally precixiouschikiren excel in productive language skills, but the salient school-relatedcharacteristic. of these children is their rapid and accurate undmIandingof verbal material. Children identified as gifted on the basis of their IQ scot cs arealmost certain to be superior in their understanding ofspoken and written language because IQ tests areheavily loaded with receptive language items (Oiler & Kyle. 1978). More variability is found in gif tedchildren's produc- tum of language since IQ testsusually do not measure the quality of children's speech or writing,

Receptive Language Skills and Giftedness Ironically, most of the research on gifted children'slistening comprehension relates only to bright students who are notperforming well in the classroom. There appear to be no statisticscomparing high-achieving 1;5 Academically Talented Children 157 gifted children's listening comprehension with their reading comprehension. Barbe and Milone (1985), however, report that there is adiscrepancy between gifted children's intellectual level and their readingcomprehension. This discrepancy between intellectual ability and readingcomprehension seems to decrease as children progress through elementaryschool (cf. Fox, 1983; Stanovitch, Cunningham, & Feeman, 1984; Whitmore,1980). By the time they are in high school, most gifted students showonly a small gap between intellectual abilityand redding comprehension. One reason that reading comprehension typically keeps pacewith older children's intellectual development is that once childrenlearn how to read (a phenomenon sometimes referred to as -breaking thecode"), they can take Control of their ownreading achievement. Children can read in an erratic fashion, skipping from one topic to another, followingtheir current interests, and still improve their vocabulary and readingcomprehension. However, the increase in the correlation of intelligence andreading cinnprehension may mean that among children with serious readingprob- lems, a decline in IQ scores occurs. Reading achievement is itself animpor- tant source of intellectual growth for bothgifted and nongided students. Since IQ tests are normative comparisons, older studentswith reading difficulties may not fare well in the comparison.

Gifted students' reading achievement.Many gif ted children are early readers, and they maintain consistent superiority in readingachievement (Durkin, 1966). This observation is true, in particular, ofgifted chikiren who are accelerated. Durkin (1966) found a smaller discrepancy between IQ scoresand reading achievement levels among early readers who were acceleratedthan among early readers who weremaintained in a chrmiologically determined grade placement. By the end of the fifth grade. someaccderated early readers in Durkin's study were reading at the ninth-grade level. Another sample of bright students with IQ scores of 120 orabove presents a more representative picture. Asseventh graders, half of these students scored two grade levels or more above their gradeplacement, and a little over a quarter of the groupscored at the tenth- or eleventh-grade level on a test of paragraph comprehension (Tyler, 1962). Evengifted children who are referred for testing because of suspectedlearning problems tend to have above average reading achieven.ent. Asignificant proportion (25 percent) of one sample of such children read two or nuwe grades above their grade placement level (Fox. 1983).

Interest in reading: then and now.Early studies (e.g.. Ilollingworth. 1926; Stedman, 1924; Strang, 1963: Terman, 1925; 'Ferman & Oden, 1947) found that gif ted children were not only better readers than other children, but that they were more interested in reading, read more widdv,

1 66 158 Academically Talented Children and read more often than other children. Records of children's reading habits over a two-month period showed that gifted seven-year-olds read more books than unselected children up to the age of 15 (Terman & Oden, 1947). More than other children, gifted children tended to select classics, biographies, nature books, fairy tales, folk tales, and legends for pleasure reading (Terman & Oden, 1947). It seems likely that this selection is at least partly the result of family values. One recent study of gifted children found reading to be the favorite pasttime of 75 percent of the group. The preferred reading material of this sample was science fiction (Van Tassel-Baska, 1983). These students could probably be considered to be highly gifted, since in the seventh grade they scored at or above the mean for college-bound seniors on the SAT-V. Tidwell (1980), on the other hand, reported that TV, dancing, sports, and talking on the telephone were the preferred free-time recreational activities in her sample. She noted that the 1,600 gifted tenth-grade students (mean IQ 137) in her sample spent an average of 11 hours per week watching television, slightly more than they spent studyMg. Since 1947, when Terman and Oden published the data reported above, TV has probably reduced the time gifted children spend reading. Nonetheless even Tidwell's sample read, on the average, 3.5 books per month, a figure that substantially exceeds the national mean. Terman and Oden's (1947) observations may still accurately reflect the annparative amount of reading done by gifted students. Moreover, if Van Tassell-Baska's (1983) report is correct, more reading or greater interest in reading could be associated with the highest levels of academic aptitude. Highly gifted students may read more and profess greater interest in reading as a pasttitne than other gifted students do.

Speech and Writkn Composition of Gifted Students Despite the paucity of research to cletermine precisely how gifted children's spoken or written language differs from that of other children, many observers con ler gifted children's speech and written language to be superior. For example, gifted children are credited with verbal behavior which is expressive, elaborate, and fluent (Renzulli & Hartman, 1971). 'Hwy are also said to have superior facility in expression (Syphers, 1972). Gilled students' debating and speaking ability are reported to be the skills that most distinguish them from other students in the classroom (Terman, 1925). All of these are probably true at least to the extent that such characteristics make it more likely that these students will be referred for testing for possible placement in the gifted program. Statements about the expressive language characteristics of this group are, however, based primarily on informal observation (e.g., anecdotal in-

f; 7 Academically Talented Children 159 formation and teachers' or parents' ratings) rather than onobjective measurement. '1.here has been little study todetermine specific quantitative or qualitative differences in the expressive languageperformance of gifted students (Bruch, 1975). On the whole, the smallnumber of studies correlating 1Q and language production lend only weak support tothe belief that high-lQ students demonstrate superior expressivelanguage skills. A large usage vocabulary is one of the most common meansof comparing the productive language of school-age children; but this andother characteristics generally associated with verbal precocity, such ascomplexity of sentence structure, do not necessarilycorrelate with higher quality language production (see, e.g., Labov, 1972).

Oral language of gted students.Studies of school children's oral language have found that, compared with their classmates,children who score high on verbal tests show somedifference from other children in the frequency of using tentative explanations, language mazes,and figurative language (Loban, 1967). They also show some differencesfrom other children in the maturity of their syntax and in theirfrequency of making grammatical errors (j. Jensen. 1973). In studying the language of dementary-schoolchildren, Loban (1967) found that children who were verballyproficient used language to express tentativeness morefrequently than did other children. 'lentative statements include suppositions,hypotheses, and conditional statements. Children, and adults as well, exhibit many false starts,repetitions, and hesitations in their ora) language. Loban studied thesephenomena, tailed "language mazes," and fOund that elementary studentswho scored high on verbal aptitude tests exhibited fewer mazes in their orallanguage than did other children. Loban (1967) also noted that very few elementary studentsused figurative language. In his study, the incidence offigurative statements was so low even among verbally proficient children that the differencebetween groups was not statisticallysignificant. Of the few children who used figurative expressions, however, all were high scorers onverbal intelligence tests (Loban, 1967, p. 56). Later research by Paivio (1971) onthe same topic suggests that within the groupof verbally precocious children there is a subgroup whose language is characterized by frequent use ofimages, sitn- des, and metaphors. This subgroup may represent those children who are interested or talented in writing poetry or fit tional prose.

Written composition.1 f, as some researchers believe, written language reflects cognitive development moreaccurately than does oral language (Greenfield, 1972), then wecould expect the written language of gifted children to be more mature than (1) their oral language and (2)the written language of other children. Neither this belief nor itsimplications

1 f; S 160 Academically Ta loard Children have been verified. According to Fearn (1981, p. 26), the beliefthat "gifted learners are uniquely equipped for writing well is a widelydisseminated generalization that enjoys not a shred of evidence." It must be admitted that the evidence is slim. Onestudy even reported the surprising result that the writing of anabove-average group of students was more syntactically mature than that of a gifted group(Dupuis & Cartwright, 1979). Dupuis and Cartwright comparedtheir gifted sample's scores on a measure of syntactic maturity (mean lengthof independent clauses) with those of Loban's (1967) above-average group,which had a mean IQ of 116 (on a group test).They were surprised to find that the above-average group outperformed the gifted group. Dupuisand Cart- wright attributed these unexpected results to the rural environmentof their gifted students. One of the few factors found to distinguish the writtenctimposition of gilled from nongifled students is the gifted students' use of rarewords and jargon (rendarvis, 1983). Gifted seventh graders wrote sciencefiction sw- ries that contained infrequently used words such as "disdainfully,""mage," "galumphing." "turbos," and "scanner." The average seventhgraders did not use jargon so extensively, and theircompositions contained fewer rare words. Strang (1956), on the other hand, found gilledadolescents' corosi- Lions to be different in both content and form fromthose of average students. The essays of the gifted students she studied contained more frequent references to abstract concepts such as morality, peace,and freedom. In addition, their essays tended to be lengthierand more fredy written than those of other adolescents. Penmanship as a confounding variable in composition.In the preschool through primary years, one of the academic problemsfacing gifted children is the extreme discrepancy between what they canunderstand and what they can write. Even early readers typically write muchless well than they read. he compositions of young students are so affected bythe physical demands of the handwriting task, that it is difficult to judge thequality of their ideas by reading their compositions. It is clear,however, that gifted children's skill in handwriting depends more on theirchronological age than on their intellectual development (Ferman &Oden, 1947). Five-year- olds who read like eight-year-olds, t..nnprehend verbal comepts like ten- year-olds, still print like other children in kindergarten orfirst-grade (Roedell. Jackson, & Robinson, 1980).

Foreign Language Aptitude Gifted children seem to learn jargon (Pendarvis, 1983)and formal grammar (Durden. 1980) withrelative ease. These characteristics suggest that gifted s:udents are likely to succeed in foreignlanguage studies. Re-

f; Academically Takata Chairra 161 search shows that in general, they do. The successof individual gifted students is not, however, insured. Characteristicsother than high IQ are often associated with students' success in learningmodern foreign languages (Stern & Cummins,1981).

Verbal intelligence and foreign language achievement. Areview of the literature on intelligence test scores, verbal aptitude test scores,and measures of foreign language competenceshows significant, though moderate, correlations among the variables. The correlationsreported in Pimsleur, Mosberg, and Morrison's (1962) review of research onjunior-high, high- school, and college-age students are typical of othersin the literattire. The correlations average about 0.45 (Pimsleur et al., 1962). This magnitude of correlation means thatIQ accounts for about 20 percent of the variance in fineign languageachievement. Verbally loaded IQ tests such as the Stanford-Binet, forexample, are more closely correlated with reading comprehension. They accountfor 40 percent to 50 percent of the variancein reading comprehension. Apparently, foreign language achievement requires someskills not measured by IQ tests.

Other correlates of foreign language achievement.As noted previously, students with high verbal IQ scores are mostlikely to be precocious readers. We know less about their skills in listening,speaking. and writing. Sonic children whose foreign languageperformame is deficient have scores on tests of vocabularyand verbal reasoning that equal tlmse of foreign language achievers. Such children oftenperform less well, however, than achievers on measuresof auditory ability (Weeks, 1974). Success in foreign language may be based on speakingand listening comprehension rather than op other verbal abilities (Weeks.1974), The ability to decode, store, and encode phoneticmaterial in an auditory mode may be a specializedskill that is relatively independent of intelligence as measured by IQ tests(Stern & Cummins, 1981). Carroll (1963) found the ability to encode phonetic materialrapidly and accurately to be a better predictor of foreign language achievementthan were IQ scores. A gifted student with auditors' processing problems mayexperience difficulty in fineign language study. The difficulty maybe compounded for such students by instruction in foreign languages thatrelic's heavily on speaking and listening exercises. Children with serious expressivelanguage weaknesses will probably experiencedifficulty with such instruction (Weeks, 1974). Although they may learn to read theforeign language. without special assistance they are unlikely to learn to conversefluently. St udents with auditory processing problems are lesshandieapped in classical language courses or in modernforeign language programs that do tmt rely on a stronglyconversational approach (Curtin, Avner, & Smith. 1983), 1 7 162 Acadenkally Talented Children School Responses to Verbal Precocity Assuming that schools should develop advanced achievement in students with exceptional potential, there are two groups who should be considered for placement in programs fbr verbally talented students:

high-IQ gifted students, and students who score high on tests of %erbal aptitude or achievement.

Both of these groups have, in general, demonstrated verbal precocity sufficient to justify exceptional programming in the verbal domaM. The first gtrup is typically regarded as verbally apt and in need of accelerated programming (Gaug. 1984). The second group is made up of apt students who may Imt score in the highest percentiles on an IQ test, but who do score in the highest percentiles on anaptitude or achievement measure of language comprehension or production. The same language arts program provided to high-IQ studentswill usually be suitable for these students as well. In fact, some of these verbally gifted students will demonstrate achievement that surpasses that of somestudents in the high-IQ group.

Identifying verbal precocity.Mere are few tests specifically designed to identifY the potential of verbally precocious school-agestudents. As the construct of intelligence evolves, however, such tests may evolve (cf. (ardner, 1983; Sternberg, 1977). For the time being current achievement can serve as a convenient proxy for specific apfitude tests topredict future achievement. F)r younger gifted students, subtests of familiar individually-administered tests such as the Peabody Individual Achievement Test (PIAT; Dunn & Markwardt, 1970), the Woodcock Reading Mastery Tests (WRMT; Woodcock, 1973), and the Woodcock-Johnson Psychoeducational Battery: Tests of Achievement (Woodcock-Johnson, 1977) are suitable. Horodetkv and labercane (1983) caution against using the criterion- referenced tests that accompany basal reading series to predict reading performance. I'hese researchers suggest that an informal reading inventory (1R1) is a better predictor than the tests associated withbasal reading series. The est of Early Reading Ability (Reid, Hresko, 1983) might also be more appropriate. It is a norm-referenced test developed specifically to assess the early reading achievenwnt of preschool or kindergarten children. The tests listed above have few items difficult enough to assess the abilities of older students. For bright children older than 10 or 12, the ceilings of these tests are too low. One group achievement test battery that ran test the tipper limits of many older gifted students'achievement ir, the Sequential "I'ests of Educational Progress (STEP) of the Educational Test- ing Service, Cmperative *lest Division, 1956-1972. 171 Academically Talented Children 163

It is important, however, not to use grade-level achievement tests for the purpose of identifying academic precocity. Grade-level achievement tests, such as the California Test of Basic Skills (CTRS), measurestudent mastery of grade-level material. Such tests haverelatively high floor and low ceiling effects. Unlike the STEP, they do not test a range of knowledge over many levels. For older gifted students, off-level testing may also be a good alternative. In this case, identification of precocity will usually involve administering college entrance tests (e.g., the SAT). These tests are given to apt students who would not normally be considered old enough to take them (Stanley, 1981). Placement.Educational policy needs to be formalized at the state and district levels to promote a consistent approach to identification and placement issues. Such policy slmuld comprise the following broad features:

1.the use of valid anti reliable assessment procedures, 2-a systematic means of considering and makingplacements, 3.the availability of program alternatives.

The educational placement of verbally gifted students in many states occurs in the context of special education procedures.Many of the states that have mandated gifted programs have included gifted education with other special education programs. In these states, changes in gilled diil- dren's programs are often subject to the state's regulations that derive from Public Law 94-142, The Education of All Handicapped Children's Act of 1975 (cf. Mitchell, 1981). Placement procedures for children identified by virtue of their high aptitude or achievement test scores (i.e., the second of the two groups listed previously) may not need to be as complex as those for high-IQ students, who may require special provisions to address precocious achievement in mathematics and other subjects as well. The multi- disciplinary placement process is, ;..)w..er, one orderly way to handle complex placement issues (cf. Pfeiffer, 198(1; Ysseldyke. Algozzine, & Mitchell, 1982). Program characteristics.The keys to mounting a successful program seem to be the following:

I.identification procedures that relate reliably' to program content. 2.program content at students' instructional levels, and 3.cognitive press.

Children must be identified with tests that reflect performance Unreliable scores guarantee poor decisions. Evaluators must have c on- fidence in the scores used to make decisions. 172 164 Academically Talented Children The tests must also relate to the content of the program. Creativity tests, for example, would be inappropriateidentification measures for accelerated reading programs, advanced grammar and composition courses, advanced literature courses, and probably for imaginative writing classes (cf. Hocevar, 1979, 1980). Programs must approximate students' instructional levels, by definition, if they are to be considered appropriate. They must move gifted students into curriculum content they would not encounter until later in the regular program. Ideally, such instruction should also match the students' pace of learning. "Cognitive press" is the instructional momentum of academic learning. Cognitive press in gifted programs implicates a comparativelyrapid pace of learning. Even underachievinggifted children can proceed through the curriculum 50 percent to WO percent faster than average children (Fearn. 1982; cf. Hollingworth, 1942). Clearly, good programs for verbally precocious students cannot carry on acceleration and exclude enrichment inthe process. Likewise, appropriate enrichment cannot be carried on withoutacceleration. Verbal materials (e.g., plays, poems, novels, histories, technical writings) that areappropri- ately advanced will, via the meaning of the text, diallenge and enrich students' thinking and ways of viewing the world. Many commentators have noted the need fOr both acceleration and enrichn. .at in gifted programs (e.g., Clark, 1983; Gallagher, 1985; Hol- lingworth, 1942). A careful distinction needs to be made between these two approaches because the selection of one approach to the exclusion of the other will strongly influence the emergence of' program options. This task is difficult in the verbal domain, however, because of the way in which more advanced texts imply morecomplex thought. Arce ler- awd or advanced content inevitably exposes students to a richer world of meaning and to the more complex, more adult issues of life.

Acceleration, enrichment, and instructional level.Although sonic educators interpret enrichment to mean enhanced curriculum at astudent's grade placement level, such enrichment is not appropriate for the verbally precocious student. First, the term "precocity" leaves no doubt of students' readiness for advanced instruction. Second, enrichment should imply provision of additional materials at the student's instruthonal level, rather than at the student's grade level. These observations contrast somewhat with reports of regular classroom teachers about their own practiceswith respect to verbal precocity (Gaug, 1984). Gang reports that the suburban teachers in her survey (n 116) unaninumsly endorsed enrichment activities and that two-thirds endorsed the use of higher level reading texts. About half of' the teachers reporwd that they were teaching high-ability reading groups. All teachers

1 73 eteadenikally Talented Children 163 activities, but fewer with high-ability groupsclaimed to use enrichment than half of these teachers (orone-fifth of the total sample)actually used of [the with their high-ability groups "atextlmok designed for a later part class'si grade level or a highergrade level" (Gaug, 1984, p.373). Grade-level enrichment wasprobably the only modificationmade for believes that this alxmt half the high-abilityreading groups. Gang (1984) situation is typical, in spite ofthe fact that researchsubstantiates the bene- students. Her fits of early reading instructionand accelerat;un for gifted report of the researchis succinct (see Box 6-2).

BOX 6-2Reading Accekration Wilkins, 1936) Earlier research (Adams 8, Ross, 1932;Dohan, 1948; McCracken, 1960; academic achievement showed thatgiftedchildren who were accelerated had greater than comparably gifted children whohad not been accelerated. More recent programs, the Program for the Astor Program (Erlich, 1979), theRapyht Project (Karnes, 1978), and of acceler- Academically Talented Students (Barthe,1980), have also shown the success provides data in favor ation programs that includereading acceleration. Durr (1981) Trezise (1978) genr:ral learningcharacteristics of of reading acceleration. As reported by and levels of the gifted would indicate that giftedreaders are ready for certain kinds instruction earlier and can movethrough material faster. students who begin One important finding relates to thefuture reading achievement of of instruction, reading in or before kindergarten.Durkin (1966) found that after six years lead over classchildren who had read before enteringfirst grade still maintained their first grade. mates who were of the same mental agebut who did not begin to read until school. Re- These children, however, were notspecifically accelerated in elementary experience as search by Clark (1979) and Callaway(1970) also supported early reading beneficial.

(cado+Realms: `tow( v ii nun Gaug, Si ;19841 "Reading AtteIvration jnt I nnt hment in thy f lymentdrv InternationM Reading Asuit iation.) 'teat hyr, i7 14i, 172474. Reprinted with permission ni the

Programming cautions.Teachers need to ix awareof the sorts of difficulties that can interfere witheffective programming torverbally precocious students. They canwork to overcome difficulties withacceleration programs by planningcarefully. Problems with enrichment programsmay be less amenable to change.however. 'Teachers often do nothave control over the structureand content of part-timeenrichment (i.e., pull-out) programs (Cox etal., 1985); and seldom withinthe regular classroom can a consistent program of enrichmentbe provided. 'There are too manyother demands on the teacher. Acceleration has not been verywidely practiced since the early1950s. Gifted education seems to be inthe process of changing educators'mituls on the subject,however. 'Teachers should be awareof some issues they are

7 4 166 Academically Talented Childyne

likely to encounter in arranging accelerative options in the verbal domain (cf. Durden, 1980; Gaug, 1984):

1.Work in accelerated placements should carry credit. 2.Grade-level, skill-based mastery testing in reading will tend to restrict accekr- ative placements. 3.Teachers may be concerned about the fate of students who complete the prescribed reading series -too soon." 4.Teachers may perceive the need to protect younger students from advanced content in literature and the humanities. 5.Students younger than 12 or 13 may experience difficulty in interpreting literature; they will encounter this difficulty, however, with lxith advanced material and with grade-level material.

When acceleration is arranged, credit should be awarded for work completed, or requireme.us should be waived for competencies demonstrated. A third grader who completes the sixth-grade reading book should not be considered to need reading instruction much longer. To place such a child in the sixth reader of another series would be, in effect, to undo the previous acceleration. Classroom teachers sometimes believe that gifted students must demonstrate full mastery of grade-levet skills before acceleration can be considered to be appropriate (Gaug, 1984). I 1, however, reading is more than the sum of subskills (cf. May, 1982), then this practice may be unjustifiable. Apt readers are especially able to assimilate many skills by inference, even though they may be unable to complete criterion-referenced skill exercises according to the typical schedule of instruction (Clark, 1983; Gallagher, 1985). Gaug (1984) notes that even teachers who endorse acceleration in reading are concerned about what to do with students who complete a textbook series early. The concern is legitimate, and it should be dealt with in a way that accommodates acceleration, not in a way that prevents it. Part of the reluctance to accelerate the verbally gifted will come from a desire to protect children from the content of reading at a more advanced level. According to Gaug (1984, p. 372), the teachers in her sample claimed "that stories at a higher grade level would be inappropriate, involving situations beyond students' emotional and social level." Terman (1925, 1947), of course, reported that gifted children typically chose difficult works for independent reading. A final caution concerns the type of' work suitable for accelerates in verbal programs. Durden (1980, p. 36) notes that younger students do not function well in "the framework of interpretative thinking." ln Piagetian terms, interpretative thinking is probably a function of the formal operations stage. Carter and Ormrod (1982) indicate that gifted children (IQ

1 75 Aradeatitaily Talented Children 167

130+) typically attain the formal operations stage aboutthe age of 12 or 13. Gifted students older than 12 or 13 might. therefore, showincreasing aptitude for interpretative thinking. As indicated above, the term enrichment has often beenused to indicate supplementary activities at grade level.Such supplementary activities are appropriate for all studentsin regular classrooms, when time or instructional method permit (cf. Guskey, 1984). For that very reason,however, these activities are notspecifically appropriate for gifted children. Since the instruction most appropriate for gifted children matchestheir instructional level, appropriate enrichment fbr them must also matchthat instructional level. Some enrichment techniques have not proven themselvescapable of promoting student growth outside the context of an academicdiscipline (e.g.. Gordon, 1961; Torrance, 1965; Williams, 1972).Indeed, such techniques were not intended to provide all essential instruction.When they are used to provide the onlyspecial programming for verbally precocious students, their value is limited.

Trends for the future. Too little has been done to provideappropri- ate acceleration and enrichment in theverbal domain. Durden (1980) reports that many of the disciplines his program(PVGY) treats are lacking in the school curriculum, even at the secondary level. Foreignlanguage instruction, which may be especially suitable for at least a subsetof verbally precocious students, may be inadequate even in high school (Cox et 1985; di Sizer, 1984). Increased concern about gifted students overthe last decade, however, indicates that perception of the need toprovide programs is growthg. The Program forVerbally Gifted Youth and the reading programs mentioned by Gang (1984) havealso demonstrated that not only is the need palpable, but that someththg can be done to meetit.

MATHEMATICAL ABILITY

Like verbal ability, mathematical ability accounts for a large part ofgeneral academic ability. Some experts. in fact, believe that mathematical abilityis the chief component of intellectual ability because of mathematics'heavy reliance on logical methods; factor analysts have shown that mathematics tests are strongly related to g (Spearman'sgeneral ability factor). Commentators in many countries have noted themilitary and industrial motive for discovering and educating mathematicalability. The pres- sing nature of this presumed need has periodicallyattracted strong support for the education of gifted students in the United States(Tannenbaum, 1981). 176 168 Academically Talenkd Chi Pfren Mathematical Skills Krutetskii (1976) distinguishes mathematical skiffsfrom mathematical

By skills and habits we mean specific actions within anactivity that a person does at a comparatively high level . ..Abilities are psychological mails on which depends an easy mastery of skills and habits in anactivity (Krutetskii. 1976, p. 71).

According to this view, skills are characteristics of anatlivity. Their existence does not depend on aparticular human subject. Goals and objectives, for example, describe skills. Abilities, on theother hand, are characteristics of a person. Abilities exist to varying degrees in theminds of students. Because academic' abilities are developed in accordwith existing knowledge. however, it makes most sense to definemathematical ability in the context of mathematical knowledge. TIUs knowledge can be described either functionally,in terms of the way it works (e.g., inparticular mathematics courses). or structurally. in terms of how it is pieced together (e.g.,in categories of skills). Structural descriptions represent mathematical knowledge in a waythat is more pertinent to the definition ofmathematical ability. (Box 6-5, however. provides a functional descripticm ofschool mat hemafic courses.)

Calculation.Skill in calculation consists of mastering basicnumber facts, such as simple sums, differences,products, and quotients, as well as the algorithms used to calculate more complex sums,differences, products, and quotients (Resnick & Ford, 1981). Calcukttionis necessary in most mathematical work, and in the United States it is thecentral skill taught in arithmetic. In higher mathematks (sometimesreferred to as "abstract math") other skills are more important, butcakulation may develop habits of attending to oumerical relationships. Apprehensionof numerkal relationships is especially important in learning algebra.trigonometry, and calculus (Edge & Friedberg. 1981: Krutetskii, 1976). Calculation, however, is not important to a conceptualunderstanding of mathematics. A major goal of the mathematicscurricuhim reforms of the early I 960s (i.e., -new math") was to impress thisfact on educators in the elementary schools. 'The attempt was unsuccessfulfOr a number of reasons, and in the United Statestoday a disproportimate amount of time is still spent on drill and practice of calculationskills. Nevertheless, whatever reforms are instituted, acertain anumnt of class time will have to be devoted to assuring students' mastery of calculationskills (basic facts and algorithms). These skills are so necessary for suucess inhigher math that the ability to compute should be a firmlyestablished, nearly automatic. habit (Resnick & lord, 1981). Academically Takated Childrem 169

Concepts.As a result of the axiomatic method of mathematics, mathematical concepts seem to grow out of themselves. For example, thecon- cept of a rational number (i.e., a fraction) is derived as the ratio of two integers. The concept of negative whole numbers can be seen to result from the redefinition of addition as subtraction, or to be the set of counting numbers and their opposites. This final example illustrates nicely thepo- tential fbr complexity in mathematical concepts. The notion of integers (positive and negative whole nurnbers and zero) is derived from the bask facts and operations of arithmetic, but the ways it is derived (redefining addition as subtraction, or conceiving of opposites)are related to each other and to concepts in higher math. In this way mathematicalcoticep- tions form a complex web of meaning. As students learn new concepts, they are expected to be able to relate them to previously learned concepts. Though some of these relationships can be noted during instruction, many cannot. In this sense it is correct to say that students -discover" mathematics for themselves (Bruner, 1960; krutetskii, 1976). Box 6-3 is a brief annotated reading list ofsources for learning more about mathematical skills and concepts.

BOX 6-3More about Mathematical Concepts

To learn more about mathematical concepts, consult the following works: finitINUMFF, W. (1968). Mathematics: the art of reason. Boston: D.C. Heath. This older liberal arts mathematics text treats mathematical concepts rather broadly, The text is accessible to ,..r.vone with a knowledge of elementary algebra. DAvis. P., & HERSH, R. (1984 The mathematical expeOent v. Boston: Houghton Mifflin. A series of Lisely connected essays that examine topics of philosophical importance. Its treatment is provot ative. Kt4,31E rskil, V. 11 97h). I he psyt holm, ol mathematic al abilities m sc huok hddren, J. Teller (trans.). I.Kilpatrick & I. Wirszup (Eds.). Chicago: University of Chicago Press. (Original work published 1968.) The nature of mathematical c oncepts is forcefully illustrated in 23 series of 79 tests contained in Chapter 8. Wirszup commends the problems to mathematics teachers. Mittfit, C., & HURIN, V. (1986). Mathematical ideas (4th ed.). Glenview, IL: Scott Foresman. This more recent liberal arts mathematics text is more comprehensive than Berlinghoff, but it requires more algebra.Itis, however. filled with interesting facts, illustrations, and problems. It would make an excellent source booi for an elementary gifted program.

Logic.The logkal method id mathematics typically involvesa step- by step deductive reasoning process in which the truth ofan assertion is understood to be conclusively denumstrated. Mathematical logk links given facts to cmiclusions by reference to axioms, definitions.or previously proven assertions (theorems). Prospective mathematicians are expected to achieve consummate mastery of this sort of reasoning. All students, however. are expected to follow such proofs andare presumed to profit intellectually from such activity (Krutetskii, 197(i). 1 76 170 Acadmingily Talented Cei !dm Textbooks typically present the concepts of each branch of mathematics principally through such logic. The deductive skills of mathematical logic, however. are not useful simply in proofs. They are essential in most mathematical schoolwork. For example. when students in trigonometry derive trigonometric identities (e.g.. sin2x = 1 cos2x). they apply such skills. The nmnerous techniques of integration (in calculus) also drawheavily on deductive skills. Application of mathematical logic is of course based on an underst.inding of mathematical concepts.

Problems solving.The solution of word problems is often believed to require particular skills. Such skills are perhaps best conceived as skills of translation (Bloom et aL, 1956). In the United States we often teach students to translate ...erbal statements into their mathematical equivalents in order to solve word problems. Skills of mathematica: translation, however, are only part of mathematical problem solving conceived more broadly. In other parts of the world, "problem solving" is the term used to characterize all mathematical activity (cf. Kruteiskii, 1976). Thus, problem-solving skills may also be viewed as the interaction of concepts, calculation, and logic.

The Characteristics of Mathematical Ability Most research on mathematical ability has devoted greater effort to identifying it than to describing it. For almost 30 ye. s, however, U.S. mathematicians interested in curriculum development have been aware of Soviet investigations into mathematical ability. Mathematicians and educators (e.g.. Heid, 1983; Wirsz.ip, 1981) note that this work is critically important to any discussion of the nature of mathematical ability. The following discussion is based on the comprehensive monograph by the Soviet psy- chologir Vadim Krutetskii (1976).

Krutetskii's study.The Psychology of Mathematical Abilities by Krutetskii (1976) is particularly valuable for its close observation of the mathematical work of approximately 200 schoolchildren who were divided into four groups: very capable, capable, average, and incapable. The very capable and capable students were considered to represent paragons of mathematical talent. Teaching experiments were constructed to allow observations of students during mi.: hematical activity, and the observed differences were interpreted as reflecting differences in mathematical ability. The experiments, which resembled the teaching experiments of Hollingworth's era (t f. Hollingworth, 1926, 1942), investigated three stages in the solution of mathematical problems. These stages were (1) gathering mathematical information. (2) processing mathematical information, and (3) retaining mathematical information. Categories of problems were devised to investi- Aeadentically Talented Children 171 gate each of the three stages of mathematical activity; one category also investigated hypothetical types of mathemati,ability. In all, approximately 500 items were used in the study. See Box 6-4 for expmples of the problems used by Krutetskii.

BOX 64 Examples of Krutetskii's Problems

FROM SFRIES I [PROBLEMS NITH AN UNSTATED QUESTION]: Before the end of a day there remains four-fifths of what has elapsed since the day began. (Unstated questionl "What time is it nowt"(

FROM SERIES VI [SYSTEMS OF PROBLEMS OF DIFFERENT TYPES]; Therearesome rabbits anci some hutches. If one rabbit is put in each hutch, one rabbit will be left without a place. If two rabbits are put in each hutch, one hutch will remain empty. How mar- rabbits and how many huwhes are there? (Classified as an arithmetic item; answer: three hutches. four rabbits.

FROM SERIES IX [PROBLEMS ON PROOF];

1, Can it be said that(-0always expresses a negative number? (Easiest item in first seven items of the series.l 2. Prove that for any a the expression (a I )(r1 3fla 4)0.1 + IP will be a positive number. [Most difficuh item in first seven items of the series.] 3. The diagonals of a traperoid divide it into four --;angles. hove that the triangles adjacent to the lateral sides are equal in area. [Classified as a geometry nem.]

FROM SERIES XXIII [PROBLEMS WITH VARYING DEGREES OF VISUALITY IN THEIR SOLUTIONS]:

1. How much does a brick weigh if it weighs 1 kg plus halfabrick? (Classified as an arithmetic item; see also 'mathematical cast of mind," which follows.] 2. In an isoiceles tri;--,gle one of the medians divides its perimeter into two parts: 12 cm and 9 cm. Determine the s des of the triangle. ICIassified as a geometry item!

Mathematical ability according to Krutetskii. The reported characteristics of' mathematical ability are organized around the three stages of mathematical activity. Altogether there are eight such characteristics, six of which describe the processing of mathematical information by mathematically able students. All eight characteristics are described and illustrated below.

1. Tb ability fur formalized perception of mathernatiall material, Jar graAping the formal structure of a problem. This ability characterized the first stage of mathematical activity, gathering information about a problem. Krutetskii and his colleagues noted that capable and very capable mathematics students apprehended very rapidly what a problem was about. This included facility in distinguishing superfluous from essential information, being able to identify missing information, and sensing how to retrieve missing information from given information. For example, ilium hearing a distance problem (two trains leaving from 180 172 Academically Talented Children

two cities at different speeds) with missing information, a capable student asked immediately if the trains were traveling in tne same direction or in different directions. 2.The ability for logical thought in the sphere of quantitative and spatial relationships, using number and letter symbols; the ability to use mathematical symbols as a mode of thought. This characteristic typifies the second stage of mathematical activity in general. Tly following characteristics are more specific. 3.The ability for rapid and broad generalization of mathematical objects, relations, and operations.The specifics of mathematical problems represent abstractions in various ways. Very capable and capable students find these abstractions easily during mathematical activity. For example, when learning abcmt negative numbers, a capable student may easily understand subtraction as the addition of the opposite of a number. 4.The ability to curtail the process of mathematical reasoning and the system (4 torte- spontling operations; the ability to think in curtailed structures. With very capable and capable mudents, steps in the reasoning process seem to fall away. How- ever, when asked, capable students are able to supply the complete sequence. Capable students need far fewer examples than average students to achieve mastery (Krutetskii, 1976, pp. 263-275; Stanley, Keating. Sr Fox, 1974; Stanley, 1981). This characteristic. which Krutetskii found striking, permits capable students to learn very efficiently. 5.Flexibility of mental processes in mathematical aaivity. Flexibility is the characteristic necessary to devise or investigate alternate routes to a solution. Capable students. for example, wcre able to elaborate multipk solutions of a problem, whereas average and incapable students had difficulty finding more than one (cf. Torrance. 1984). Flexibility is extremely helpful in overcoming obstacles in the solution of' a problem; it permits students to shift from one strategy to another. With capable students, the discovery of one correct solution does not seem to suppress discovery of cnhers. 6.Striving for clarity, simphrity, economy, and rationality of solutions. This characteristic- addresses the goals of mathematical elegance and scientific parsimony. For example, upon finding one solution to a problem, capable and very capable students often spontimeously revise their work with a view to parsimony and elegance. ("Parsimony" and -elegance" refer to the related qualities of sim- plicity and clarity in a formal proof, solution, or scientific theor0 7. The ability for rapid and free reversal of the mental process in mathematical reforming. This characteristic allows capable students to reconstruct the process of solving problems and to see the patterns and connections of' mathematical (cm- cepts. For example. a capable student who was studying complex numbers was asked to multiply (a + bi). The student noted immediately that the result provided fOr the factorization of the sum of two squares because (a + bi)(a hi) a4- b2, a fat tin itmion not accounted for in elementary algebra. but analogous to the convenient factorization of the difference of two squares. As noted in our discussion of skills, many mathematical concepts are related in this way. S.Grnecalrz,rd memory of mathematical celatiowhips, typecharacteristics. schemes at-

gument$ and prmp, methods af problem solving,and print110.e.:11 itappittat-h. bis is ability for the third stage of mathematical activity. Capable students demonstrate both longer and better organized mathematkal memory than other students. The generalized or abstract nature of this characteristic is its significant feature: it is not a memory fOr details. For example, although a very

S Academically Ta leased Children 173

capable student remembered the essential features of problems nine months after working on them, the student began to forget concrete details by the end of a lesson, and inesscntial details were rarely remembered at the end of a lesson. Incapable students, however, tried to remember everything about a problem.

Finally, the study describes one characteristic that was studied only in capable and very capable students. Krutetskii (1976, pp. 314-315) implies that only in capable students is the operation of this characteristic observable, because its observation requires the presence of very well developed skills. This characteristic, mathematical cast of mind, manifests itself in four ways. These four types of developed mathematical ability can he represented as a 2 x 2 cell. (See Figure 6-1.)

Analytic Geometric

Harmonic Analytic Analytic type 8

Harmonic Geometric type A Geometric

FIGURE 6.1Types of Mathematical Ability among Capable Students (mathematical cast of mind) According to Krutetskii, all mathematical ability is characterized by a verbal-logkal (or "analytical") component. and therefore the type of mathematical ability "can only be a matter of the greater or lesser development of the visual-pictorial [or 'geometric] component. Accordingly, otw can speak of the predominance of the visual-pictorial component over the verbal-logical only in a relative sense" (Krutetskii, 1976, p. :316). The capable student with an analytic type of ability demonstrates verv Atrong verbal-logical performance combined with weak visual-pictorial performance. Such a student's spatial concepts are weak. The student does not feel the need to use visual supports and uses an analytic approach even in problems designed for visual-pictorial solution. The capable student with a geometric type of ability demonstrates abovr-average verbal-logical performance combined with vrry 5trung visual- pictorial performance. Such a student's spatial conceptse very gym!. "1-he student feels the need to use visual supports and uses a visual-pictorial approach even in problems designed for analytic solution. Capable students of the harmonic types A and B possess both Atrong verbal-logical skills and Atrong visual-pictorial skills. Their spatial concepts are good. The harmonic type A student, though able to use visual supports in solving mathematical problems, does not benefit f-rom their use! After solving a problem, usually in the analytic mode, such a student can, how-

1S2 174 Academically Tainard Childrrts ever, construct a visualrepresentation. The harmonic type B, by contrast, benefits from the use of' visual aids in solvingmathematical problems. Krutetskii (1976) reported that the "brick"problem cited in Box 6-4 distinguished reliably between analytic and geometric castsof mind. In solving this problem, students classified as geometersdrew pictures and diagrams. Analysts reasoned or composed equations.

Quantitative traits not part of mathematical ability.Several traits associated with mathematical activity were excludedfrom Krutetskii's structure of mathematical ability.These included (1) rapid working tempo;(2) computational abilities; (3) memory for symbols,numbers, and formulae; (4) ability for spatial concepts; and (5)visualization ability. The first excluded trait (rapid working tempo)should not be roninsed with a rapid ability to apprehend a problem(the first characteristic of mathematical ability), or with a rapid ability togeneralize mathematical material (the third characteristic of mathematicalability). This observation implies that power tests may be more effectivethan speed tests in identifY- ing mathematical ability. The exclusion of computational abilities mightsurprise some teachers. Krutetskii does notintend that students remain ignorant of basic facts or algorithms. 1k notes,however, that very good calculation skills are not symptomatic of mathematical ability. This observatim isextremely important to teachers of the giftedbecause of the emphasis placed on calculation in elementary classrooms in the United States. The third excluded trait represents the influenceof row memory in mathematics. An average memory is adequate; even a poor rote memory need not diminishmathematic-al ability. This observation reinforces the notion that mathematical ability concernsabstractions more than par- ticulars. The fourth awl fifth excluded traits derive fromKrutetskii's observations of the mathematical cast of mind of capableand very capable students. Apparently, even geometry can beunderstood with weak visual- pictorial ability, whereas weak analytical abilityseverely limits mathematical ability, and these limits cannot be overcome evenby strong visual-pictorial achievement. This observation also points up theabstract character of" mathematical abilit y.

Gender and mathematical ability.A number of studies substantiate the better performance of I 1-year-old males onthe SAT-M (Benhow & Stanley, 1980; 1982; Stanley & Benbow, 1983). This differenceis reflected in significant mean score differences (one-half toone-third standard deviation) between the scores of males and females, aswell as in the greater numbers of males scoring at extremely highlevels (Benbow & Stanley. 1980; Stanley & Benbow, 1983; Stanky etal., 1974). Krutetskii (197(3) also

1 S3 Accidenivolly Takatted Children 175 noted the greater number of males in his small samples of very capable students (brilliant 7 to 10 year-olds). For a time the cause was thought to be poor spatial skills among females, but this hypothesis has been shown to be weak (Boles, 1980; Kru- tetskii, 1976). Differences in achievement and course-taking are now being investigated (e.g.. Eccles, 1985; Maines, 1985). The SAT, of course, explicitly confbunds ability and achievement (i.e., aptitude is commonly understood to be a compound of achievement and ability). One recent study of mathematically able students (Weiner & Robin- son, 1986) found that males' SAT-M scoresaccounted for about twice as much variance in the mathematics course-taking as did females' SAT-M scores, whereas the males' SAT-V scores accountedtbr none. On the other hand, the researchers found that females' SAT-M and SAT-V scores accounted for equal amounts of variance in course-taking. This finding may be related to Krutetskii's notion of mathematical cast of mind.

Scht.ol Responses to Exceptional Mathematical Ability Krutetskii's study illustrates just how the characteristics of mathematically able students allow them to learn more rapidly thali average students. These characteristics confirm the need to alter the curriculum for mathematically able students. School resimnses to exceptional mathematical ability can he categorized as identification efforts, eligibility determination and pia:ement decisions, and instructional programs that involve acceleration and enrichment. The general observations about the establishment of school programs that were noted in the verbal precocity section also apply to mathematical ability. They will not be repeated here.

Identification.Exceptional mathematical ability is a matter of degree, and it is determined most reliably by performance onnorm-referenced tests. Unfimunately, reliable individually-administered comprehensive tests of mathematical ability do not exist. The mathematics portion of the Scholastic Aptitude Test (SAT-M), validated by the Study of Mathe- matically Precocious Youth fin- the identification of mathematical brilliance. is of limited use. "fhe characteristic that suits it for the identification of brilliance, many difficult items, makes it unreliable for distinguishing average students from exceptional students (i.e.. those whoseperformance in mathematics is approximately 2 standard deviations above the mean). The lack of a comprehensive individually-administered test of mathematical ability makes it impossible to identify mathematical talent on the basis of mathematical ability. The best alternative is to base identification on mathematical achievement. Ideally, achievement tests for this purpose would

1 84 176 Academically Talented Children sample the skills that most reflect the characteristics ofmathematical ability: (1) apprehension, (2) logic, (3) generalization, (4)curtailment, (5) flexibility, (6) elegance, (7) reversibility of thought, and (8)retained insights. They would not be influenced strongly by (1) speed, (2)computation, (3) rote memory, or (4) spatialskills and concepts. Available individually-administeredtests approximate thesere- quirements to different degrees. The mathematics clusterof The Wood- cock-Johnson Psychoeducational Battery, Part II: Testsof Achievement (Woodcock & Johnson, 1977), is perhaps the best all-purposechoice. The mathematics subtest of the Peabody Individual Achievement Test(Dunn & Markwardt, 1970) might also be used, though it is shorter andprobably less reliable than the Woodcock-Johnson mathematics subtest. One test in particular should be avoided. The arithmeticsubtest of the Wide Range Achievement TestRevised ( Jastak &Jastak, 1978) is not suitable because it samples computational skills exclusively; itis also timed. Exceptional mathematical ability cannot be inferred from a highlevel of computational skill (Krutetskii, 1976). Both the Woodcock-Johnsonand P1AT subtests sample skills more pertinent to mathematicalability. Typical group achievement tests ofgrade-level performance should also be avoided for identification and placement purposes. They maybe used as a source of referrals. (See Howley etal., 1986, pp. 26-32, 54-57 for more extensive discussion.)

Placement.Eligibility criteria for gifted programs should include exceptional mathematical ability (operationalized for the time being ashigh achievement). Some school districts are apparently pursuingthis option already; placement in most districts, however, is contingent on ahigh IQ score (Cox et al., 1985). Teachers of the gifted therefore need to be alert for signs of exceptional mathematical ability among both high-1Q children and amongother children as well. Almost always, exceptional ability in mathematicsis associated with above-average, though not necessarily exceptional, IQ,The con- verse is also true: Gifted childrenalmost always prove more capable in math than average students. Teachers of the gifted also need tobe aware that a few gifted students will exhibit characteristics thatindicate weak ability in mathematics. Enrichment. The need to alter the content of programs to accommodate capable mathematics students has long been recognized.The mathematics curriculum revisions of the 1960s effected permanentchange at the secondary level. Many of the high-school textbooks now in use owe a large debt to the pioneering work of groups at Yale University andthe Univer- sity of Illinois. (See Box 6-5 for a descripti of secondary mathematics courses.) At the elementary level, however, many teachers wereunable to implement the "new math.- The elementary mathematics curriculumis IS5 Academically Talented rhildren 177 therefore still based on computation, and mathematicsinstruction adheres slavishly to this outmoded curriculum (Kersh &Reisman, 1985). The extreme literalness of elementarymathematics instruction re- presses the achievement ofall children, but it is especially harmful toable students (Hersberger & Wheatley, 1980; Howley etal., 1986; Kersh & Reisman, 1985). In other countries more timeis devoted to coaching students in manipulating concepts and insolving problems (Travers & McKnight, 1985). The tests on which Krutetskii (1976)based his observations, for example, contained "arithmetic" items thatresembled word prob.. lems from elementary algebra texts in the UnitedStates. Krutetskii's subjects usually solved these problems without recourse toalgebrak. symbols and equations. The complete revision of the mathematicalcurriculum ought to be an essential goal of curriculum development at theelementary level in the immetilate future. Past experience i»dicates that elementaryteachers, however, must be more mathematicallyable and more involved in active teaching. At present most elementary teachers do notfeel competent in the subject (1Cersh & Reisman, 1985). If such curriculumreform does not Occur On beha!f of allchildren, teachers of the gifted need to understand that curriculum reforms must be carried out in thegifted program for mathematically capable children.

BOX 6-5The Contents of High School Mathematics Courses

Arithmetic, often taught as *general math" in high schools, is notconstrued as a high school level course (cf. Wekh, Anderson, & Harris, 1982). Algebra I and II are typically not taught sequentially, but with a year of geometryintervening {Travers & McKnight, 1985). Algebra I introduces the useand development of a1gebraic expressions, linear and quadratic equations, and solving wordproblems. Algebra Ilreviews these techniques, and typically introduces conk sections,logarithms, and the concept of functions. Geometry is usually taught following Algebra!, In the United States these inn ses concern fuchdean geometry almost exclusively (Travers & McKnight,1985). They aim to teat h students to use the axiomatic method to reason logically and thinkvisually. Trigonometry introduces a family of functions different fromalgebraic functions, but derived from geometrk: notions: sine, cosine, tangent, andothers. Various applications of these functions are usually taught as well. Trigonometry mayfollow Algebra H Of be taught as part of it, Trigonometry, however, usfrsalgebraic techniques. Calculus course includes both differential and integral calculus.Differential calculus, taught usually in the first semester of the twelfth grdde, developsthe idea of limits and its application in finding derivatives for algebraic, logarithmic, andtrigonometric functions, Derivatives are interpreted geometrically as the slope of a curve.Integral takulus is the process of finding the function of which a givenfunction is the derivative. Integrals are interpreted geometrically as the area under a given curve or thevolume described by the rotation of a given curve around an axis.

Accekration.The workofthe Study of Mat hematicaliy Precocious Youth (SMPY) over the last decade has establishedthe validity of radical

1 6 178 Academically Talented Children acceleration for mathematically brilliant youth (Stanley & Benbow, 1983). SMPY has also been influential in demonstrating the need for moderate acceleration among less extremely gifted students (Cox et al., 1985). The work of SMPY establishes not only that acceleration is important, and even essential, but also that it can take the place of enrichment. Accelera- tion is able to accomplish this feat through strategies like early entry and grade skipping. Instead of suffering through unnecessary drill in computation, students who are prepared adequately can undertake a college algebra course instead of a pre-algebra course. For such students college algebra would be an enriched program.

The dilemma of school responses to mathematical ability.The aim of both enrichment and acceleration is to promote achievement. Whereas enrichment is necessary for all students (especially at the elementary level where calculation is virtually synonymous with mathemitics), the need for acceleration in mathematics is unique to students who are exceptionally able in mathematics. Cox and her colleagues (1985), however, found that even moderate acceleration (one to two years over the course of 12 years) was seldom practiced in gifted programs. Radical acceleration was rare. At the same time, the sort of enrichment practiced was not the sort described above. Instead, enrichment was practiced most often in resource programs that were uncoordinated, redundant, and ineffective in promoting achievement (Cox et al., 1985). Given these facts, the most difficult opiion seems to be enrichment, which requires changes in curriculum structure, curriculum materials, and teacher preparation throughout a school system. Acceleration, on the other hand, requires the cooperative action of a few concerned individuals. Whatever tbe school response. teachers should ensure that both high- IQ children and exceptionally able mathematics students take ample course work in mathematics. Learning mathematics requires the taking of courses (Howley et al.. 1986). Bright student:, who avoid math classes are likely to remain _gnorant of this very important subject. Student differences in the number of courses taken account for a substantial portion of the variance in mathematics achievement of high-school seniors (Welch, An- derson, & Harris, 1982). Research also indicates that arithmetic achievement affects performance in Algebra I (Noland & Michael, 1984), and that mastery of algebraic skills (Algebra 1 and II) .strangh influences perf'or- mance in calculus (Edge & Friedberg. 1984).

SUMMARY

This chapter examined various kinds of academic talent. It related the characteristics of high-1Q children to the characteristics of verbally and mathematically apt students, and it distinguished between academically 167 Acadenskally Talenkd Children 179 significant talents and talents that do not correlate very highly with academic achievement. The discussion of verbal aptitude focused on the performance of gifted children in reading, writing, and learning foreign languages. It noted that early reading achievement did not predict general academic ability very well, though about one-third to one-half of gifted children read before starting school. It also noted that gifted students' writing differs little from the writing of average children, at least through junior high school, and that foreign language aptitude seems to incorporate some skills not strongly associated with general academic aptitude. The section on verbal ability concluded with a discussion of school responses to verbal precocity, including a section on optimal identification and placement practices pertinent to all varieties of academic talent. The discussion of mathematical ability began by examining the function of mathematical knowledge of computation, concepts, logic, and problem solving. "[he important role of concepts and logic in mathematicswas noted. Krutetskii's thoughtful study of the characteristics of' mathematical ability was reviewed. The ability to generalize concepts and operate logically were anmng eight prominent characteristics noted. Discussion also included an explanation of five. traits excluded from the structure of mathematkal ability, but comnumly ascribed to matlwmatical ability by other observers. A discussion of school responses to mathematical precocity followed. It considered the need to identify on the basis of achievement and the need to revise the mathematics curriculum, in order to improve mathematics Mstruction, especially at the ekmentary level. The discussion also considered the need to accelerate the progress of mathematically apt students.

I St, Children with Talent in the Visual and Performing Arts

I. Focusing Questions II.Arts Talent, Creativity, and Academics A.Arts Talent versus Creativity 1. Creativity in arts identification 2.Creativity in arts training 13. Kinds of Arts Talent 1. Music 2.Visual arts 3. Dance 4.Drama C.The Study of Arts Talent III.Music as a Concern of Schooling A.The Pedagogical Dilemma of Musical Talent B.The Formal Musical Language 1.The syntax of music 2.Musical notation C.Aptitude in the Musical Language 1. Nature versus nurture 2.The musical environment D.Signs of Musical Talent 1.Absolute pitch 2.Rhythm 3. Melody and pitch discrimination 4.Sensitivity to harmony 5.Musicality E. School Response to Musical Talent 1. Identification

180 Childress with Talent in the Vinwl and Performing Arts 181

2, Programs 3. Acceleration and enrichment IV.Talent in the Visual Arts: The Arts Context A. The Visual Arts: A Definition B. The Media of the Visual Arts C.Determination of Visual Arts Talent 1.Drawing skill 2. Cognitive complexity .3. Perseverance D. The Measurement of Artistic Potential 1. The indicators t talent 2. Identification ( artistically talented 3. Nature versus nurture E. School Responses to Art Talent 1. Enrichment and acceleration V.Dancing and Acting as Performing Arts A.Dance and Dancers B. Characteristics of Dancers 1. Personality traits of gifted dance students 2. Physical characteristics of dancers 3. Psychomotor and cognitive skills C.The Schools' Response to Dance Talent 1, Identification 2.Programs 3. Acceleration and enrichment D.Acting and Actors 1. Personality characteristics of actors 2. Cognitive skills in acting E. The Schools' Response to Acting Talent 1. Identification of acting talent 2. Programs 3. Acceleration and enrichment VI, Summary

Focusing Questions 1.Why is it helpfuland why is it misleadingto compare the use of artistic media to the use of verbal language? 2.What is the relationship between the development of creativity and the development of arts talent? 3. Why is inadequate instruction in the arts an impediment to the identification of talent in the arts? 4.Is absolute pitch required for a high level of musical performance or musicianship? 5.What is the role of cognition in the development of arts talent? 6. Why is publi.: school training in the arts so often delivered as enrichment rather than as a well-structured curriculum?

1 182 Children with Talent in the Visual and Performing Arts

ARTS TALENT, CREATIVITY, AND ACADEMICS

The arts form a crucial part of the intellectual legacy of a culture. They provide media through which aesthetic interpretations of the world can be expressed and understood. Like verbal expression, arts media allow individuals to express and comprehend ideas. Unlike verbal expression, however, music, art, and dance do not allow for a one-to-one correspondence between a symbol and a clearly expressed meaning, as is possible in language or mathematics. The meaning of many works of art, therefore, is only partially accessible. That quality is part of what makes the arts so intriguing. In the Western world, arts media form the basis of personal statements that may become profoundly meaningful to contemporary or future audiences. Because of the combination of' personal and universal elements in artistic symbol systems, learning to understand art and to express oneself in an arts medium is more difficult than learning a language. Language learning entails exposure to and practke with an idiom whose usageby comparison with expression in the media of artdepends almost totally on convention. One cannot say, "See the cat," and, in fact, mean "See the fish." On the other hand, one's use of a bold red line in a painting does not by itself mean anything in particular. The meaning depends on the a next of' the particular painting and On artists' previous uses of bold red lines. The line does not denote something specific; rather, it may connote a web of related associations within a context that depends on the painter's aesthetic sense, the history of painting, and the knowledge of the viewer. Because arts production involves a highly personal use of a medium. many arts educators believe that instruction should involve freeing the creative impulse rather than channeling it. These educators (see e.g., Lowenfeld & Brittain, 1970) maintain that creativity in the arts occurs naturally in all children. They believe that arts talent will develop only if' it is freed from the constraints of convention. This view reinforces the idea that creativity in general is the goal of arts instruction. Creativity, however, cannot be the goal of arts instruction, because creativity does not function apart from a particular medium. Al- though it is essential to encourage creative vision so that individt,als can produce art, the encouragement of creativity cannot guide the development of arts talent. The creative impulse is a kind of motivation. Students need to be motivated to learn any academic subject. As they are successful in the subject, their motivation for further learning will, in theory, increase. The same is true of instruction in the arts. As students learn to control an arts medium and to understand its legacy, they will become more creative. For these reasons this chapter views the development of arts talent as a legitimate academic issue. It suggests that ars talent is learned and that 191 Children with Talent in the Visual and Perfmning Arts 183 proficiency in the arts should be taught. Early exposure to or instructionin the arts often precedes later achievement in an art form. However, preschool instruction is not essential for talent development in the arts. As an important part of the intellectual tradition of Western civilization, arts understanding (or literacy) should be the minimum goal for all educated individuals. Mastery of expression in one or more art forms should be an instructional goal for those who show particular aptitude for the arts. Most of this chapter is devoted to a discussion of children who seem toexhibit exceptional taient in music, visual arts, dance, and drama. The beginnings of arts talent in children are examined and the kinds of instructionsuch children need are briefly indicated. Each sectien also reviews some ofthe basic concepts without which any discussion of these arts is impossible.

Arts Talent versus Creativity In Chapter 4 we noted that creativity was initially defined in terms of arts talent. There we suggested that arts talentoffered a more fertile and intellectually relevant field for instructional efforts than creativity training. Here we want to make sure that creativity is not mistaken for talent in the arts. This point needs to be made because creativity tests are,unfortunately, among the most frequently used methods of identifying artistically talented students (Clark & Zimmerman, 1984a). Creativity is viewed as a global cognitive skill, like intelligence in its scope. It is unlike intelligence, however, in the wayit works, in its measurement, and in its social implications. For instance, ample researchshows that children's scores on creativity tests can be improved rapidly (Perkins, 1981; Rose & Lin, 1984). Whether this is due to the to ease with which creativity can be trained, to the efficiency of training programs. or tothe inadequacy of creativity tests has not been settled.

Creativity in arts identifration.Part of the desire to identify the arts talents of' children comes from the perception that schools are doing little to nurture such talents. Many persons interested in thiseffort are, however, discovering that it is very difficult toidentify arts talent. There is no consensus on the natural developmental sequenceof' skills in each art form, nor on the sorts of pedagogicalmethods that are best. Moreover, most schools do not have complete arts curricula; the arts are typically taught in enrichment programs. Considering the limitations of typical arts programs, even subjective assessment of arts talentwould be difficult. Under these circumstances, it may seem that any effort at identification is better than no effort. Hence, less than serviceable methods of identification come to be used. Because we still think quite naturally of artists as bring creative, some school systems have adopted creativity tests to identify students talented in the arts (Clark & Zimmerman, 1984a). 184 Chi letrwe with Tam lthe Viseq god Pstforwisw Arts As a primary selection mechanism, creativity tests have limitations so severe that they are useless for identifying specific talent in the arts. First, they were not designed to identify arts taleni. Torrance's original intention was to account for academic achievement not predicted by IQ tests (Tor- ranee, 1984). Second, students selected by creativity tests are presumed to be creative individuals, but creativity tests do not specify the field in which their creative potential might express itself. As Perkins (1981) notes, almost any human activity under any circumstances can be considered creative. Be- cause creativity has been so broadly defined, there is no way of inferring from creativity tests whether or not a student has any kind of arts talent. This J. identification of students using creativity tests implies that such students will be instructed in creativity training programs. Identifying and training creativity will not, by itself, develop arts talent. Development of arts talent depends on knowledge of technique in the arts. Creativity training programs are, unfortunately, more readily available than arts training programs.

Creativity is arts training.There is no reason why c,eativity (i.e., divergent thinking) exercises cannot fo n a part of instruction in the arts. As classroom calisthenics they make good instructional sense when th 7 are relevant to the task at hand. Good arts teachers have used just exercises for same time, in pan because arts teachers are themselves divergent, nonconformist personalities (Perkins, 1981; Getzels, 1979). This observation illustrates the error inherent in applying creativit) research to the development of arts talent. Divergence is characteristic of artistically talented students. As a trait separated from the exercise of craft, it is merely a personality trait, and it does not need to be taught to such students.

Kinds of Arts Talent Important distinctions exist between understanding, performance, and creation. The distinctions apply in different ways to each art form.

Music. When people think of musical talent, they probably think most often of the ability to play a musical instrument. This is an example of talent for performance. It sometimes happens, however, that quite talented performers do not understand the music they can play so well (Shuter, 1968). Such performers are said to demonstrate poor musicianship. More- over, even virtuoso performers (who by dermit ion exhibit exceptional musicianship) do not usually compose music. Possibly, composition is another kind of musical talent. Some training in performance is a prerequisite for a high level of

193 Children with Mks' in the Vaned and Petfoneiv Arts 185 musical understanding. Nonetheless, when a school identifies musical tal ent, someone needs to give thought to whether instruction will consist of training to play an instrtunent, to compose, or to understand music. Choos- ing to train exceptionally talented students in performance, in particular, may require extraordinary resources to locate and employ talented teachers. Training in composition will be a rare choice, since, as noted above, the sort of curriculum necessary to support this activity is lacking in many schools.

Visual arts.Performance is not a substantial feature of the visual arts. Objects of art are viewed after the act of creation (with a few avant- garde exceptions). The visual arts involve making objects of art or understanding art objects made by others. Understanding the visual arts may seem less technical and forbidding than understanding music. The literature of visual art is, however, both oldet and more extensive than that of written literature, and far older than the musical literature (which dates only from about 900 A.D.). Arts understanding implicates at least 5,000 years of history: The serious visual arts student is working in a tradition in which the technical possibilities of virtually every medium have been elaborately (if not completely) explored. As in the history of musk, each new historical period changes artistic conventions in some way. Iconography (the symbol system used by works of art), composition, media, and techniqae may all be affected.

Dance.Dance talent includes both performance and choreography (composition), although performance talent is most commonly identified. Virtually all instruction in dance consists of training to perform. Dance creation (choreography) is usually the work of retired dancers or of dancers who decide they would rather choreograph than perform. Understanding dance requires some sort of physical training, just as understanding the visual arts requires some experience making works of art, and understanding music requires some experience in performance. The history of dance, however, is much less accessible than the history of the visual arts or music. Only recently has a system of "dance notation" been developed to record choreographic ideas. In choosing to identify and train students who are talented in dance, emphasizing performance skills over choreography and dance history seems justified.

Drama.Talent in acting is a performance capability that depends on the effective delivery of words, gestures, and movements. It is closely associated with an understanding of the history of the theater and interpretation of the dramatic literature. Therefore, dramatic talent requires knowledge of and interest in literature as well as the ability to portray characters in a convincing manner.

194 186 Children wish Talent in flut Visual and Perforsprim Arts

The Study of Arts Talent Arts talent has been studied by artists, who have described their own creative efforts. It has also been studied qualitatively and philosophically by aestheticians and critics. Psychologists and educators have studied the arts in a more descriptive and quantitative fashion. The autobiographical reports of artists allow for insight into their methods of working and motives for creating (see e.g., Morgenstern, 1956). Biographies of artists also provide an understanding of the personal histories and theoretical viewpoints of individual artists. Aestheticians and critics have also discussed the nature and results of artistic creation. Usually aestheticians have considered issues such as the nature of beauty, the uses of art, and the origin of the artistic impulse. Critics have evaluated the art products of particular artists in the context of the arts legacy. Understanding the literature of criticism and the philosophy of art requires extensive knowledge of critical and philosophical discourse, as well as of the particular art works or types of art work under discussion. Such an approach to the study of art is quite scholarly. Artists, aestheticians, and critics have been writing about art for centuries. Within our century, however, psychologists and educators have also begun to study and write about the artistic process. These social scientists have been concerned with a variety of issues related to arts production. Cognitive psychologists, for example, have been concerned with the processes of seeing (in the visual arts) and audiation (in music). Developmental psychologists have been interested in tracing the sequence of arts skills acquired by both normal and exceptional individuals. Educators have studied arts talent in the school context. They have considered the value of arts instruction in general education as well as the rationale and methods for instructing talented young artists. Because this chapter is concerned with pedagogically relevant findings about arts production, it relies more heavily on studies by educators and psychologists than on biographical or critical works. In particular, the chapter emphasizes studies that relate to the identification, development, and training of artistically talented youngsters.

MUSIC AS A CONCERN OF SCHOOLING

Music now surrounds us almost without interruption. Americans love music, and they use it to improve retail sales, citizenship, and academic skills. They understand the power of music, but they do not understand its meaning (Barzun, 1965). According to Allen Britton, Dean of the University of Michigan's School of Music, "music lin the public schools) can never be juq

105 Chi !drew with Talent is the Visual and Performing Arts 187 music, an art that is its own justification; it is always being asked toproduce 'something more' such as better health or citizenship, improved reading or math skills" (Murphy, 1980, p. S-16),

The Pedagogical Dilemma of Musical Talent Most children get a bit of music on a regular basis in the elementary schools. From middle school through senior high school most schools sponsor a band ,,,nd a chorus, which typically serveonly 10 percent of the high- school popula.:on (Petzold, 1978). Instruction on instruments that are not band instruments I. '.!strMg and keyboard instruments) is much more rare. The sort of music insaaction provided in school does not favor theidentification or cultivation of exceptional musical ability. At the same time that more people hear more music than everbeibre, most of what most people hear is music that is merelyserviceable. Such music exists only to influence trade. As a result, many people have difficulty distinguishing between musical trash and music that is good of its kind. Barzun (1965) characterizes as "fanatical zeal" the devotion withwhich people cling to the music they like. This devotion is enough to cause most adults (including teachers) :o regard wide differences in knowledge as mere differences of taste. The classical music literature, which contains much music that is good of its kind, remains a foreign language to most students (cf. Miller,1979). Few high-school general music classes require students to listen tochalleng- ing music (Miller, 1979). Western "classical" musk is perhaps not thebrAt music in the wor!ti, but it is very complex, it is profoundlyexpressive and meaningful to our culture, and, like any great literature, its relevance en- dures, In our culture it is imperative that very apt music studentsbecome fluent in the Western classics. In fact the Western classics arequickly becoming world classics, played in Delhi, Tokyo, and Nairobi, aswell as in New York, London, and Vienna.

The Formal Musical Language In order to understand what musical talent is, readers mustunder- stand a few things about the formal musical"language." It is helpful to make the analogy between the musical symbol system andverbal language in order to explain a realm of meaning with which some readers maybe unfamiliar. The distinction between the cognitive and emotional import of music is difficult to grasp. It requires some sense of the waymusic works. Like all analogies, the analogy of music and language ismisleading. Music does not convey meaning in the same way asverbal language. Partic- ular phrases and combinations of harmonies do not inevitablycommuni- cate specific emotional -:ontentin a universally understood dialect (Barzun,

196 188 Children with Takut the Waal mid Perfentsiag Arts 1965). Psychologists of music (e.g., Jackendoff & Lerdahl, 1981) persist in misunderstanding this fact, perhaps because their primary musical activity is listening (Murphy, 1980). Nonetheless, music has a kind of syntax. and it exists in varieties that are analogous to dialects (Sessions, 1979). Concerted mtellectualeffort is necessary to understand complex musical works. Thiseffort is certainly pleasurable, but the artfulness of music does not consist in its presumed emotional content.

The syntax of music.Music, like speech, depends for its effect on the passage of time. As with speech, extracting as muchmeaning from the passage of time as possible requires attention toseveral qualities simuhane- ously. The informed music listener understands that the music of a particular time or place is organized (much like verbal syntax) in conventions of rhythm, harmony, and melody that combine to produce a characteristic effect. Those who master the syntax of several such "dialects" manage to make accessible to themselves a large literature of meaning. The basic qualities of musical syntax are melody, rhythm, and harmony. Mehdy, a sequence of pitches joined together,is in some ways the most basic quality of music. Perception of melody, however, depends on exposure to, if not training in, the musicalconventions of one's culture. The lack of ability to hear more than one type of melody is. however, a major impediment to the musical growth of many people (Murphy, 1980). Rhythm refers to the temporal pulse of music and, in its broadest sense, implicates tempothe speed at whichmusic is played. At a more specific level, meter establishes a fundamental pulse pattern (usually in groups of two or three beats) against whichrhythmic elaboration can occur. At the most specific level, rhythm may refer to a specific rhythmkal motif, a pattern that can be used in one or several meters tohelp structure a very elaborate piece (e.g., much of Beethoven's fifth symphony is structured on the opening four-note motif.) Harmony is the most complete aspect of musical syntax. It is the interaction, according to prescribed conventions, of two or more pitches played simultaneously. In any harmonic pattern, there exist both horizontal pitch (melodic) and vertical pitch (chord) arrangements. Vertical arrangements may themselves proceed horizontally (i.e.,in "chord progressions"). In its simplest form, harmonic perception refers to perceiving the interval between two notes, to identifying chords, and to anticipatingchord progressions. This skill is among the last to emerge in children (Shuter. 1968). There is a more difficult harmonic skill, however. In complex compositions the listener must simultaneously perceive both vertical and horizontal harmonic changes, each associated with characteristic rhythmical Children with Talent in the Visual and Performing Arts 189 motifs and variations. Students, however, need to know that this complex skill is worth learning, and that it can be learned (Broudy, 1958). Unfortu- nately, most never do (Shuter, 1968).

Musical natation.For the musical intellect, the invention of musical notation (aboutA.D.900) equaled the significance for the verbal intellect of the earlier invention of the alphabet (about 1700 Lc.) The system of musical notation allows complex works to be brought into existence, interpreted carefully, studied, and reinterpreted with fresh insight. Even most playing "by ear" (i.e., without training) is merely repetition of a melody with a suitable accompaniment. Complex music, however, cannot be played in this way. The accompaniment to a Bath fugue, for example, cannot be improvisedit has no accompaniment: it is atonce (and this is the incredible beauty of the fugue) all melody and all harmony. There is a natural limit to the length, coherence, and complexity ofa composition that can he improvised. Musical notation helps extend this natural limit.

Aptitude in the Musical Language The preceding review clearly indicates the degree to which the musical "language" is a cultural product. Musical precocity, far from being the hereditary possession of exactly the right equipment. is tantamount toa coincidental opportunity to make fluent and original use of the cultural heritage (Feldman, 1979; Petzold, 1978; but cf. (;ardner. 1973; Kwal- wasser, 1955).

Nature versus nurture. The nature-nurture debate has engagedmu- sicians as well as psychologists. Do certain families tend to raise musically talented children, or are such children born to talented families? Aswe note elsewhere, the evidence in favor of the hereditarian hypothesis of intelligence is not firm. The genetic studies of musical talent (e.g., Feis, 1910; Friend, 1939; Galton, 1869/1962; Kwalwasser, 1955; Mjoen, 1926; Shuter, 1964) are even more flawed and therefore less conclusive than the studies of the heritability of intelligence. Nonetheless, because musicians tend to come from "musical" families, the literature tends to confirm the hfreditary origin of musical talent (e.g., Gardner, 1973). The consensus (by no means complete) illustrates theway in which inconclusive research results often support prevailing beliefs (Gould, 1981). Like Feldman (1979), Gordon (1980), and Walker (cited in Murphy, 1980, p. S-24), we believe that the quality of the child's musical environment is critically important, whatever the child's genetic endowment.

1 .)8 190 Children with Talent ix the Visual aad Ptiforwing Arts

The musical environment.In a musical environment, both parents usually play a musical instrument and actively encourage their children to do the same. Their children repeatedly hear music that engages the par- ems' interest, and they probably take part in musical discussions. The wider the range of the parents' musical activity, the wider the children's understanding is likely to be. The recent research of Bloom and his colleagues (Bloom, 1982, 1985; Bloom & Sosniak, 1981) documents the extraordinary instructional effort mounted by exceptionally musical parents. Very musical families seem always to have known what recent researchers (Gordon, 1980; Perzold, 1978; Shuter, 1968; Wooddell, 1984) have discovered: Mu- sical aptitude can be increased if music instruction starts early (cf. Kwal- wasser, 1955, pp. 61-73).

Signs of Musical Talent Musical precocity is related to a number of special conditions and experiences. Shetler (1985) tentatively reports that umsciously manipulated prenatal experiences make a marked difference in ability even among the children of musically talented parents. Nonetheless, these conditions and experiences are in the nature of developmental owns, and caution is in order. Just as early speech is unrelated to IQ, so the emergence of a single musical omen does not usually imply great talent. A developmental pattern of early skills, in the context of musical activity, however, probably does indicate talent by the age of five to nine (,7f. Gordon, 1980).

Absolute pitch.Prthably the most interesting of' musical Omens is "absolute pitch," the ability to identify accurately any note played. Absolute pitch is a specific talent that amazes those who do not possess it. Surely, this seems to be the case of a talent inborn. Research on absolute pitch,however, suggests this is unlikely. One study reported by Shuter (1968) examined four groups of musicians (professors and students. n = 261) of' moderate to high musical ability. Within ability groups, absolute pitch subjects began theie lessons one to six years earlier than subjectswithout absolute pitch. The largest difference was in the highest ability group. In some cases, subjects were completely successful in naming notes played on the earliest studied instrument, and completely unsuccessful in naming notes played on the instrument studied subsequently (even when the instrument studied later had become their major instrument!). Absolute pitch has been thought to be necessary fbr a high degree of musicianship. Shuter (1968) reported a study that compared highly talented professional musicians gifted with absolute pitch to equally talented

199 Childress with Take: in the Visual wed ferferwing Arts 191 colleagues who did not have absolute pitch. The researcher transposed the beginnings of standard repertoire pieces and asked subjects to identify which versions were not played at correct pitch. The subjects without absolute pitch scored better than those with absolute pitch. Shuter concluded that absolute pitch, though helpful to exceptional musical achievement, is not essential.

Rhythm. A number of music educators have proposed the idea that rhythm is the most fundamental aspect of the musical syntax (cf. Mark, 1078; Orff, 1955; Seashore, 1938/1967, p. 147-148; Sessions, 1950/1965, pp. 11-15). The proposal is not, however, substantiated by research. jer- sikl and Bienstock (1935, cited in Shuter, 1968) appear to have been among the few to use other than anecdotal records. They photographed children aged two, three, four, and five who were instructed to walk or to move their hands in time to music. A sample of seventeen adults served as a control group. The researchers observed that children's accuracy improved with chronological age, but concluded that subjective impressions of children's ability to keep accurate time were untrustworthy. Whether rhythmic skills develop before, after, or concurrently with melodic skills is still an unresolved issue (Horner, 1965; Shuter, 1968). In spite of this fact, rhythmic movement activities dominate early music instruction in schools (e.g.. Aronoff, 1965; Mark, 1978; Nye & Nye, 1970).

Melodyand pitch discrimination.Some music educators (e.g.. Wing. 1954) continued all along to doubt the relevance of early rhythm and movement activities to musical training. They believed instead that a child's sense of melodic shape was the earliest musical skill to emerge. According to Shuter (1968), a good ear for melody seems to develop in two stages. At first, children perceive only the directir,;, of movement of the notes, but not the correct intervals between thel4!Tie second stage involves perception of correct intervals in a melods; . dnd this stage is not attained by all children. Shuter (1968) compiled observations made by musical parents on the emergence of the musical skills of their children. According to these reports, such children first recognized tunes and were able to reproduce individual notes between the ages of 9 months and 2.5 years. The earliest age reported by the parents for singing a complete tune correctly was 1.3 years, with the mean alxnit 2 years. Gesell and Ilg (cited in Shuter, 1968, p. 69) report the following developmental norms for average children:

using phrases. not usually on pitch2.0 yrs. recognize several tunes 3 yrs. a f ew sing entire songs correctly 4.0 yrs.

2110 192 Chi :dim with Adria im the Visual and Prtforwhy Arts Nye and Nye (1970, pp. 540-543) narrate the following approximate norms for averagi: children:

12-26 months begin to imitate pitch 26-36 months sing tunes within the intc: ..11 of a third or fourth 36-60 months individual differences emerge 60-72 months sing tunes within the interval of a sixth

Kwalwasser (1955) observed that, even though it is possible to identify the average age at which particular skills are attained, the actual age of skill attainment by different children varies greatly. In fact, Kwalwasser (1955) found that the range of achievement differences within any particular grade exceeded the range of differences between all levels in the elementary school. This finding is more likely to reflect lack of instruction in school than hereditary differences in musical ability.

Sensitivity to harmony.Harmonic patterns (tonality, simultaneous melodies, chord progressions, the interaction of melodic theme and accompaniment) are accessible to perception only after a great deal of listening. This explanation may account for the fact that although average students begin to sing songs correcdy by the age of six or so, average students do not begin to prefer concords (e.g., intervals of a major third or fifth) to dis- cords (e.g.. major or minor seconds) until the age of 11 to 13. Musically apt children probably acquire this level of understanding much earlier (Horner, 1965; Kwalwasser, 1955; Shuter, 1968; Shetler, 1985). One musical child known to the authors, for example, shr-wed a preference for thirds (demonstrated by producing them spontanc isly at the keyboard) at the age of five. This same child could reproduce individual notes at nine months and sing complete songs on pitch at age two.

Musicality.Musicality is sensitivity to musical meaning (Mursell, 1958. p. 146; Shuter, 1968, pp. 222-226), and the development of such sensitivity implicates musical knowledge (cf. Broudy, 1958). Musically talented students arereceptiveto this sort of knowledge. Some examples of such knowledge follow. The harmony of Mozart (and his contemporaries) is very much like that of our familiar homophonic music (i.e., music with a single, prominent line of melody). Nonetheless, homophonic tonality of the late eighteenth century is more rudimentary than ours. The late eighteenth century seldom used seventh chords, for example. As a result new listeners to this music find it "childlike" or "boring." As another example, the music of Bach tends to be more dissonant than the music of Moran. Bach's harmony is most frequently built on the vertical coincidence of simultaneously played melodies, a difference that

2111 Chi/drys with Mind inthe Visual and Performing Arts 193 implies differences in the elaboration of t!lose melodies and in the tonality which shapes them. In the case of Bach, listeners who have not perceived the conventions of the early eighteenth century are apt to find this music "mechanical" or "cold and intellectual." The remarks noted in these examples revrml the predicament of listeners who have not perceived the conventions, much less understood the art, of the music to which they have listened.

School Response to Musical Talent Though musicians and music educators have often claimed that schools do not take music seriously enough (Hollander, 1978; Murphy, 1980; Petzold, 1978), their observations seem to conflict with Barzun's (1965) characterization of Americans as a devotedly musk-loving people. Perhaps the contradiction can be resolved by noting that Americans make too great a distinction between musical pleasure and musical knowledge (cf.liroudy, 1958; Marcuse, 1978). Whereas we love music a great deal, we do not necessarily insist on understanding it.

IdentOcation.The schools' identification of musical talent is jeopar- dimd by the lack of programs that implement a sequential musk curriculum (Broudy, 1958; Howley et al., 1986; Murphy, 1980; Sessions, 1979). Still, some alternative methods of identification do exist. They are considered in more detail in our textbook about teaching gifted children (Howley et al., 1986). One alternative is to use standardized tests to select musically talented students. Shuter (1968) probably still provides the most extensive review of music tests. She favors the Wing Tests of Musical Intelligence (Wing, 1948), whereas we recommend (Howley et aL, 1986) the Gordon Music Aptitude Profile (Gordon, 1965). Gordon's (1979) Primary Measures of Music Audiation, for children aged five through eight, also merits scrutiny in the context of the present discussion, which stresses the importance of early training. There are other tests available as well. Although all probably measure musicality (see previous discussion) to some degree, none assesses a child's aptitude for music performance. Performance ability can be identified by jun auditions similar to those that precede admission to special schools. In this case, the instrumental instruction provided must be excellent. The program must be controlled and taught by talented professional musicians. The most typical way that schools identify some musical talent is through participation in such school music programs as do exist. In these programs (band and chorus) some students naturally emerge as more talented than others. Since so few students participate in these programs, this method is probably not effective or efficient.

211 2 194 Children with Taloa in the Visual and Perforwthig Arts

Programs.Three groups of musically capable students can be identified: (1) talented young performers, (2) students with general musical talent identified by standardized music tests, and (3) students with general academic talent. The third group has been identified as possessing strong interest in aesthetic issues (Tidwell, 1980). Many observers (Hollander, 1978; Kwalwasser, 1955; Seashore, 1938/1967, pp. 6-9; Shuter, 1968; Wing, 1955) note that musicality (if not performance ability) is strongly related to general intelligence (measured as 1Q). It is important to obsetve the above distinctions when programs are implemented. Of the three groups of students, public schools are probably in a position to serve in a single program both those academically gifted and those musically talented students that can be identified by standardized tests. These students could be expected to benefit from group keyboard instruction or group string instruction, from concurrent instruction in music history and theory, and from concurrent attendance at required listening and discussion sessions. Substantial effort, however, will be required to provide a sequential curriculum of this sort for these children. With hard work, such a sequence could be carried out by a team of talented amateurs (not volunteers) working with a knowledgeable music teacher. (For sensible remarks about the talents needed to teach arts programs for gifted students see Alexander, 1981.)

Acceleration and enrichment.Music serves a recreational role in our schools. Most music instruction that currently takes place in school should therefore be considered enrichment. Without school programs that teach a sequential curriculum, acceleration is dearly impossible (cf. Broudy, 1958). Even when students work with a good private teacher, few receive instruction in the history of music, Or any listening experience. Talented young performers, however, may be lucky enough to be referred to the few specialized arts high schools, where acceleration is theoretically possible. As Shuter (1968) observes, young children should first experience music as fun. It is unfortunate that though we like music, we persist in teaching even our most musically apt students little about it (cf. Miller,1979).

TALENT IN THE VISUAL ARTS: THE ARTS CONTEXT

This portion of the chapter looks at the characteristics of students who are talented in the visual arts. We review these characteristics in the context of certain assumptions. First, we assume that the definition of art determines those talented in art. If all representational activity is considered art, for example, then all students who engage in such activity are artists. Second, we assume that training uncovers talent. Only in an environment that

2113 Children with Talent in the Visual and Pe 'farm* Arts 195 nurtures talent can students with uncommon abilities be discovered. Fi- nally, we assume that art is a cognitive activity embedded in an intellectual tradition (see also, Arnheim, 1966). Art depends on literacy in both the techniques of expression and the history of the tradition. These three assumptions allow us to view the talent of artists as developmental. A developmental view of talent emphasizes the importance of nurture. It equates art talent with other intellectual talents and focuses on the interaction between ability and instruction.

The Visual Arts: A Definition The question, "What is art?" has been answered in various ways hy different artists, art critics, aestheticians, and art educators. Although it is not the purpose of this chapter to review all such definitions or todecide among them, we must darify some issues surrounding the definitionof art in order to consider the nature of individuals talented in art. Definitions of art include the pre-Renaissance Christian belief that art is worship accomplished through the act of creation (Chiari, 1960) and the more contemporary belief that art is self-expression(Arnheim, 1972). The earlier view emphasizes the technical nature of the artistic process, whereas the modern view stresses the psychological import of the artistic process. In practice, neither view alone supports a complete aesthetic. Consequently. art education based exclusively On either view (technique or psychology) is unbalanced. At one time art education consisted only of disciplined instruction in the crafts of drawing and painting (Duncum, 1985). Later, art education became the arena in which children expressed their feelings thi ough creative activities (see e.g., Lowenftid & Brittain, 1970). The former approach molded children's art in conformity with models of "good" art, and the latter approach validated children's art irrespective of judgments related to the quality of their art products. Both approaches, however, limited the range of children's artistic development(Taylor, 1966). A more comprehensive approach to the development of artistic talent incorporates both the training of technical skills and the nurture of individual expression. According to Aso (1984), art is experience of the world realized through various media. This experience of the world, however, is shaped by historical and cultural forces (Korzenik, 1981). Art, then, is a mediation of experience that is both direct and itself mediated by history and culture. An important force guiding the work of any artist is the artistic language of previous generations. Art is defined in relationship to its ante- cedents. Good art both uses and transcends its heritage. The artistic process is not the exact replication of the world; it is not the process of copying the works of others; and it is not an egocentric catharsis. It goes beyond both technique and individual self-expression. 2 4 198 Children with Talked in the Visual and Petfartning Arts

Without technique and knowledge of the language of art, meaningful self- expression cannot take place. Without the expressive effort of the individual, art cannot come into existence. To make art, artists must know the history of their medium; they must know themselves; they muse know how to express themselves through their medium; and they must be willing to do so. Although the training of art talent cannot produce art, it can teach craft, make the artistic legacy accessible, and encourage effort (cf. Arnheim, 1966). Skilled artists are able to use their craft in a dialogue with the art of earlier times and thereby aim to produce ar .hat is both novel and uni mrsal (see e.g., Aso. 1984).

The Media of the Visual Arts According to Herbert Read (1959), "art" is:

a means of conceiving the world visually ...The artist is simply the man who has the ability and the desire to transform his visual perception into a material form. The first part of this at:lion is perceptive. the second isexprrssive.but it is not possible in practice to separate these two processes: the artist expresses what he perceives; he perceives what he expresses, (Read, 1959, p. 12)

This definition can help us examine the ways in which visual perception can be realized in material forms. As artists seek to gain control over a medium (e.g., drawing, painting, textiles, photography), they learn the techniques and history particular to the medium. A certain amount of this learning, however, may influence their production in other media as well. For instance, skill in pencil drawing may influence aa artist's ability to paint. In general, however, the most transferable skills in art are those involving perception. The artist learns to see in a characteristic way perhaps conditioned by work in a particular medium. This characteristic vision may then manifest itself in works regardless of the medium ir which it is realized. In practice, many artists limit their attention to one or several media. Very few artists paint, sculpt, weave, and take photographs, for instance. The tendency of artists to use only a few media probably reflects some element of ability as well as an element of interest. The skills of a photog- rapher differ considerably from those of a painter. Nevertheless, when we concern ourselves with the identification of childrentalented in the arts, we are most likely to look for and to develop talent inthe most commonly used art techniques. These techniques include skills in drawing and painting, at and perhaps sculpture and printmaking as well. Drawing, however, is often viewed as the "executive skill" basic to all art forms (Stalker, 1981, p. 49).

205 Children with Talent in Se Visual and Performing Arts 197

Determination of Visual Arts Talent Determining who is talented in the visual arts would be easy if all activities involving manufacture were thought of as an. Based on this view, the most prolific producers would be the most talented. At one time, art educators perpetuated this view of art. In the Fortieth Yearbook of the National Society for the Study of Education (Whipple, 1941), art educators presented their views on art in American life. Chapter titles from this yearbook provide examples of the crmcerns of this group of educators: "City Planning," "The Domestic Setting Today," "Landscape Design," "flower Arrangement," "The Handcrafts," "Art in Industry," "Clothing and Personal Adornment," and "An in Commerce." The art educators who contributed to the yearbook failed todistin- guish between the aesthetic (or visually pleasing) and the artistic. Accord- ing to Best (1980, p. 78), an "art form must at least allow for the possibility of the expression of a cor,-eption of life issues such as moral, social, and political issues." Clothing, for instance, can be visually pleasing, but it is difficult to see how clothes can ever express anything coherent about the substantive issues of life. A more limited definition of art leads to a more specific way toidentify those talented in art. Stalker (1981, p. 50) maintains that successful artists demonstrate executive drawing skill, cognitive complexity, and a high level of perseverance. With such a definition art educators can distinguish arts talent from aesthetic awareness, a characteristic typical of gifted students in general (Tidwell, 1980).

Drawing skill.Although most children draw, only some exhibit unusual talents in drawing. According to Clark and Zimmerman (1984a, p. 17), "all differences in drawing skills . .of highly able and less :able students are differences in degree and not it. kind." Precocity and sophistication in drawing skills may characterize childrer talented in the visual arts. At the same time, however, it is mclear how to define precocity or sophistication in drawing.Clark and Zimmerman (1984b) describe a model of evaluating children's drawings on a continuous scale ranging from naive to sophisticated. They write, "applications ofthe Naive-Sophisticated model are age/grade dependent; whether a child's art work can be judged naive or sophisticated is dependent upon reference to an age/grade norm" (Clark Se Zimmerman,1984b, p. 215). The determination of sophistication in this model depends on a number of elements including the organization of a drawing, an indication of linear perspective, the use of varied line widths, and the use of shading. Another approach to the determination of precocity in drawing skill involves the evaluation of children's performance on tasks required for

2 (16 1913 Chadree with Watt in theVisual and ProfanelyArts skilled drawing. Children with high levels of manipulative dexterity, speed, and control may be viewed as potentially talented in the arts even if they do not seem to be motivated to produce art works. There is no consensus among art educators concerning the nature of talent in drawing. Some art educators, for example, regard early representational skills as an indicator of art talent. Others regard the bold use of line and color as an indicator of art talent. Whatever approach is used, however, some children possess skills in drawing that surpass those of other children. Although we can and should encourage the artistic development of all such children, we cannot determine in advance which of thinn will want to become artists. We are even less likely to determine which of them will become artists (cf. Getzels & Csikszentmihalyi, 1976).

Cognitive complexity.In general, 3rtistically talented individuals have higher than average levels of intelligence. Clark and Zimmerman (1984a) review the literature on the relationship between art talent and intelligence. They conclude, "Though superior intelligence and superior art abilities are clearly interdependent, not all children with a high IQ possess art talent. All children with superior art talent, however, do possess a higher than average IQ" (Clark & Zimmerman, 1984, p. 14). According to Stalker (19P1, p. 50), "cognitive complexity [isl the most adequate equivalent of the kind of intelligence necessary for artistic performance." She views cognitive complexity as the ability to:

(a)handle gmater amounts of information, (b)handle more complex information environments, (c)assimilate conflicting or inconsistent information, and (d) make better judgements about complex information. (Stalker, 19131, p. 50)

Perseverance.Art educators seem to agree that, in general, achievement in the visual arts is associated with extensive practice (see e.g.,Dun- cum, 1985). In most cases those who prove to beartistically talented as adults begin artistic production early in childhood (Meier, 1939/1966). Many children, however, who are prolific in early and middle childhood stop drawing in early adolescence (Gardner, 1973). It is not certainwhy this change occurs in some children, although levels of ability, interest, and encouragement by others may have some influence onadolescents' continued artistic production. It is also not certain that the most artistically prolific children are actually the most talented. Nevertheless, children who are consistently making images should be encouraged to continue to dr) so. It would be wise to consider such children at least potentiallygifted in the visual arts.

20 7 Chikirm with Tana ix the Visual and Alarming Arts 199

The Measurement of Artistic Potential For the purpose of identifying those students most talented in art, it would be convenient if there were a reliable and valid norm-referenced test of art talent. Such a test would need to be based on norms of artistic performance at various ages (Clark & Zimmerman, 1984b). A test of this sort would also have to define and measure the subskills of art talent. Unfortunately, art educators do not agree on the skills that comprise art talent. Neither do they agree that early development of art skills necessarily relates to art talent (see e.g., Saunders, 1982).

The indicators of art talent.Clark and Zimmerman (1984a) present a thorough review of the literature concerning the characteristics of students talented in art. Their review reveals a lack of consensus among those who have studied the development of talented students. For example, some researchers conclude that children talented in art specialize in drawing one subject matter, but others conclude that talented children draw a wide variety of things. It is beyond the scope of this text to examine these speculations, but we recommend that readers to whom this topic is critically important consult Clark and Zimmerman's (1984a) work.

Identification of the artistically talented.Because of the many unresolved issues concerning the nature of art talent, it is unlikely that any standardized test would accurately predict art achievement. In practice, students presumed to be talented in art are identified using various selection procedures. According to Clark and Zimmerman (1983, p. 26), the most commonly used selection procedures (listed in the order of their popularity) are (1) self-nomination, (2) portfolio review, (3) classroom teacher nomination, (4) interview, (5) creativity test, (6) informal art test, (7) art teacher nomination, (8) achievement test scores. (9) structured nomination, (10) peer nomination, and (11) parent nomination. All of these approaches, however, have limitations and the authors suggest that "procedures currently in use and those recommended for use.. be critically examined and evaluated before they are implemented" (Clark & Zimmer- man, 1983, p. 28). Since previous achievement ofte,1 correlates with achievement later in life, schools should perhaps identify and train all students who are precocious in art development, all those who are prolific in artistic production, and all those who are cognitively advanced (see e.g., Salome, 1974). This approach would assure that the students most likely to be taler ..ed in ar, would have the opportunity to develop their abilities. Such an approach, however, would depend on a very high societal regard for art production. Finding and nurturing future artists would have to be a social priority. 200 Children with Talent in the Visual end Performing Arts

Nature versus nurture.In the absence of social validation for artistic production, the nurture of artistic talent is not widespreaci. Under such circumstances, it is convenient to assume thatart talent is a genetic endowment that will mature without training and encouragement (Clark & Zim- merman, 1984b). This view provides a rationalization for neglecting the artistic training of children. it is clear, however, that training of some sort, either self-trainingor tutelage, is necessary in order for even the most talented studentsto develop fully. Gardner writes:

individual differences in genetic endowment are, at most,one contributing factor to the eventual level of artistic accomplishment. Whilea deficient hereditary history may preclude high achievement, only arduous training and development of skills can convert the potential for excellence into its realization. (Gardner, 1973, p. 231)

Bey...me we cannot predict talent in the absence of training,we are not likely to find a test that identifies early in childhood the select few who will become the most talented artists. Only ifwe train everyone, can we be sure of missing no one. We also cannot be sure that talent coupled with the most extensive training will produce artists. According to Getzels and Csikszentmihalyi:

One does not become an artist just by painting. To paint might be the only thing that matters subjectively. But to be able to earn a livelihood and to develop a self-concept as a bona fide artist distinct froma "sometime painter," artistic behavior is not sufficient. One must be legitimized by the appropriate social institutions. (Getzels & Csikszentmihalyi, 1976, p. 185)

Since this legitimation is governed by numerous social, political, and economic forces, we cannot assume that the most talented artists will be allowed to earn their livelihood through the practice of their art. Under such conditions, all of the identification and training effortson behalf of students talented in art will be compromised.

School Responses to Art Talent In elementary schools, "perhaps art is the least understood, themost difficult, and in many cases the poorest-taught subject" (Johnson, 1965,p. 55). Often what is called art instruction in elementary schools involves the completion of craft projects that are carried out almostas a manufacturing enterprise. Such projects usually conform to specifications that prohibit both the learning of technique and the exercise of imagination. Many elementary classrooms are decorated with children's attempts to imitate the teacher's Easter bunny. Santa Claus, or paper-cup Snoopy. While such

2 0 Children mat Talent ha Ma Visual and Performing Aris 201 projects may improve classroom climate, they do not succeed in teaching art to children. Another appro3ch to elementary school art instruction is to view it as a channel through which children can express their emotions (see e.g., Lowenfeld & Brittain, 1970). This approach stresses the affective import of art education and ignores the cognitive import. It discourages any attempt to provklz structured lessons that might limit children's free expression. In secondary schools, art education is beset by other problems. Gener- ally, art classes are offered as electives. Often bright children are discouraged from taking an electives, and below-average children are sched- uled into these classes. According to Silverman and Lanier (1965, p. 117), "the validity of art as a legitimate atea of study for ail students has been recognized in less than 10 percent of American senior high schools."

Enrichment and acceleration.In some school districts art instruction is part of the enrichment program provided to children who are identified as gifted. In a very small number of other districts, art instruction is provided to children selected because of their talent in art (Alexander, 1981). In either case, the instruction provided is of such short duration and of such limited scope that it can only be considered enrichment. Extensive, high-quality instruction in art necessarily provides both enrichment and acceleration. lt requires a curriculum that Iii.:ludes art history, art technique, and art criticism. Although programs based on such a curriculum are rare, they do exist. Such programs are most often found in specialized high schools that offer programs for talented students (Cox & Daniel, 1983).

DANCING AND ACTING AS PERFORMING ARTS

Singing, playing musical instruments, dancing, and acting make up the performing arts. All have been intimately connected since ancient times (Sorrel, 1967). In our educational system, however, there is a hierarchy of respectability among the performing arts. Music, including both singing and playing, is typically considered the most academic of the performing arts, and most public schools make a feeble effort to teach basic music appreciation and performance skills. Of the other performing arts, acting and dancing, acting is usually considered the more academic. Most educators, at least grudgingly, accord to acting the status of an art, but many view dance as a form of sport (cf. Best, 1980). These attitudes affect talent development in the performing arts. School priorities determine that children will have little opportunity to develop their musical ability in the public schools and even less opportunity

210 202 CM ldran wit* Talon t ithe Visual and Poribruciag Arts to develop ability in drama or dance. Lack of support fordrama and dance is partly due to the prevalent view that the arts are noncognitive and therapeutic, rather than academic. Drama and dance, however, are given even less support in the schools than the other arts,perhaps beg luse they are more overtly :eductive than music or the visual arts.Performers are often seen as instrumental, rather than essential, and as technicians, rather than artists. Singers, musicians, actors, and dancers are,it, a sense, media (Arnheim, 1972). Their bodies and behavior are components of the art work. Their appearance and behavior are the center of attention. In a successful performance, the appearance and behavior of the performing artist pleases the audience. In fact, the audience will be unable to distinguish between the person of the artist and the role assumed by the artist in a successful performance. The difference between creative and interpretive skills is an issue in the performing arts. It is not in the literary or visual arts. Only a small proportion of artists in the performing arts originate works. Composers, choreographers, and playwrights, like painters, sculptors, and novelists, create their art with relative autonomy. In contrast, thepublic nature of the peforming artist's work has implications for both the status of the performers and their talent development opportunities.

Dance and Dancers Dance is the least academic of the arts because of the lack of an adequate system of notation with which to compose and record dances. Musical notation specifies in great detail the qualities of the music; lyrics and scripts give the content of songs and plays even though they do not specify interpretive vocal inflections, gestures, and facial expressions. Dance notation was, until this century, virtually nonexistent.Dances were passed down from one generation to another, like folk tales. No collected bad-, -f- dance works is readily available for study, criticism, and stimulus to furthedevelopment. There are no basic works that student master in order ?t,. learn the history of dance technique and content. This lackof a written record prevents dance from achieving the academic statusthat characterizes other artistic disciplines. Dance is, in a sense, a form of acting; however, the discursive content of a dance is necessarily much more limited than that of a play. Bet4-`,1Seit employs words, drama makes much more specific statements thandance. Although dances often have narrative content, dance movements are primarily connotative. The more denotation demanded by the content,the more dance resembles pantomime. The primary vehicles of expression in dance are positionand movement. Usually dance employs stylized movement, butordinary movements, 2 Children with Talent he the Visual and Performing Atte 203 such as walking, are not uncommon components, particularly in dances choreographed by contemporary minimalist artists (e.g., Yvonne Rainer). The main difference between movement in dance and movement in other activities is that all movement in dance is intended to achieve an aesthetic effect. The varieties of dance (social, folk, theater, and concert) differ in their relative emphasis on aesthetics, recreation, and pantomime, however. Social dances are designed primarily for recreation rather than for performance. In contrast to choreographed dances (i.e., incidental dance in plays or musicals or classical concert dance), social dances are neither physically demanding nor complex. Folk dances usually combine elements of &ming for recreation and dancing for an audience. The steps in folk dancLs are prescribed by tradition and are descended from ritualistic dances that enacted community events and myths (Sorrel, 1967). Folk dancing is often more complex and more strenuous than social dancing. Despite these differences, the basic psychomotor skills of dancing are present in all forms of dance; and identification of talented social or folk dancers would be a viable means of selecting students for formal training. Such selection would need to be accomplished fairly early in children's school careers, however, becanse professional dancing demands many years of training. Characteristics of Dancers Successful dancers have usually begun to study dance seriously by the time they are fifteen years old. Girls often begin training at a much younger age. According to some ballet critics, dance instruction for ballet must begin by the time the child is eight or nine years old (Kirstein, 1983). Like musicians, dancers make the decision to become professional artists early in life, usually between twelve and eighteen years of age (Ryser, 1964). By their high-school years, most female dancers are spending a large portion of their free time in dance classes, at least an hour and a half per day (Braunschweig, 1981). Outstanding achievement in dance, as in other artistic and academic disciplines, demands early and continuing commitment from both child and parents. Twyla Tharp's early dance education is typical of tint of many successful dancers:

Although Twyla Tharp grew up in a small town in California, her mother was determined that she should have a proper education in music and the arts. "She had the luck," Tharp says, "to find these teachers who came out of the Ballet Russe tradition...Later she drove hundreds of miles to take me to study with Beatrice Illeauchampi Colleneue (a student of Anna Pavkwa)." (Vaughan, 1984, p. 54)

In additiun to commitment, accomplished dancers share several personality traits, physical cha acteristics, and cognitive skills. 212 204 Chi kW* with Minot in the Visual and Psiforstirg Arts

Personality traits of gifted dance students.Personality traits of dance students resemble those of other artists and art students (Wilson, 1964). Good dance students value aesthetic and philosophical ideas; they also take pleasure in arts other than dancing (Ryser, 1964). They are perfectionistic and willing to devote considerable energy to dance (Alter, Denman, & Barron, 1972). The emotional intensity common to creative artists is also found in dancers' self-descriptions. Both male and female dance students see themselves as highly emotional, but male dance students describe themselves as less confident and subject to more internal conflict. This conflict may be caused by sex-role stereotypes that proclaim dancing to be a feminine activity. Dancers share the childhood isolation that seems to chwracterize the lives of many artists. Particularly during adolescence, they feel alienated from peers (Ryser, 1964). This feeling may arise from their different interests as well as from the need to spend much of their out-of-school time in dance classes. As a group, dancers tend to reject many middle-class mores. Ryser (1964) views dance students' rejection of middle-class values as a result of their isolation during adolescence. Whateve ,. the cause, a similar rejection is characteristic of most artistically creative groups. Nonconformity to middle-class standards of behavior is typical of artists. The comments of one of Ryser's subjects portray an attitude prevalent among creative subjects:

I was always different from the people I met in high school .They were doing all the things that kept them from thinking. (Ryser, 1964, p. 112)

Feelings of superiority over the general population are also typical of artists (Maddi, 1975), and dancers are apparently no exception. lt is interesting to note that dancers also disdain other dancers (Ryser, 1964). The dancers studied by Ryser saw other dancers as ill-informed, too specialized, and uninterested in other arts. Each saw him or herself as an exception to this generality, however. Ryser suggests that this degree of idiosyncrasy may also characterize other types of artists.

Physical characteristics of dancers.Although modern and jazz dance have somewhat less stringent physical requirements than ballet, there are some physical characteristics common to dancers in allthree major forms of concert dance. Dancers must be slender, well-proportioned, flexible, strong, and well-coordinated. The relative importance of some of the characteristics varies with the sex of the dancer. Strength is important for both, but more important for male dancers, at least in ballet. Flexibility is probably more important for female dancers whatever the form of concert

213 Cadres with Takist 11th. Visited and Performing Atte 205 dance. The following characteristics are commonly cited as standards for the most strictly defined body type, that of the female ballet dancer:

Balanced body proportionsThe length of legs and arms is important. Short legs cannot achieve impressive extensions in arabesques and kicks. On the other hand, a disproportionately short torso sometimes lacks the strength to sustain high arabesques. Short arms are not so lyrical as long arms, but arms that are too long may look grotesque on stage. Strong feet with aflexibk archThis is important because so many female dance roles are performed en pointe. Even with support in toc 31suc3, this form of dancing is extremely difficult if the dancer's feet are weak or inflexible. Pelvic structure that allous "tuns ourMost ballet steps are based on dance positions that require the dancer's legs and feet to be turned out in a nearly 90 degree angle to the front of the body. Some women's pelvic bones angle in a way that limits their ability to attain good turn out.) Sloping shouldmSquare shoulders are usually less flexible and sloping shoulders give the appearance of a long neck. Following the influence of Balanchine, nearly all American ballerinas today have a slender, elongated appearance. (Jacob, 1981)

More important than these characteristics, even for female ballet dancers, are the psychomotor skills used in dance. Nonetheless, psychomotor skills are related to body structure, and postural anomalies or over- weight disqualify even the most graceful dancer from a stage career in the United States.

Psychomotor and cognitive skills.Among the psychomotor skills required for dancing are (1 ) ability to discern and imitate kinesthetic patterns in detail, (2) quick memorization of those patterns, and (3) superior reten- tion of learned kinesthetic patterns. When choreographers are working out a dance, they may change the dance 20 or 30 time.. (McKayle, 1966). Dancers who cannot quickly learn new combinations are at a distinct disadvantage, and their chances for success are limited. The ability to discern and imitate complicated body positions and movements is essential. I f dancers had to have each element of a position or movement called to their attention separately, learning a new dance would be a slow, arduous process. The ability to see and imitate even the most subtle component of a dance step or combination of steps enables the talented dancer to progress rapidly. This skill is learned, but the skills of discriminating small differences in movement and of coordinating complex movement patterns in imitation of a model takes years of training and practice.

Ilike all of thew characteristics. this quality is a t onsitkration, but not eswntial to success; Anna Pavlova was considered to have rather poor turn out IFonteyn. 1979).

214 206 ChildrenWith Mada the Visual and Pelona* Arts Interpretation requires understanding of the semantic content of the dance and of its emotional tone. Interpretation in dance requires someof the same skills as acting; sensitivity to gesture and its significance, sensitivity to facial expression, memory for both,and the ability to imitate them. Musical intelligence adds to the dancer's interpretive ability (Duncan, 1927). Dancers' interpretive ability is closely related to their style, an important variable in dance. Although technicalability is necessary to achieving eminence in dance, the most acclaimed dancers are not always the most technically competent. For example, Isadora Duncan's contribution to dance was not based on her technical competence somuch as her style, which was so dramatically different from ballet that it constituted a new form of dance (Kraus, 1969). Style reflects both dancers' anatomical makeup and their personalities. The fluidity with which a dancer moves, the height of leaps, andthe precision with which the dancer defines movements are amongthe elements of individual style. These areprescribed primarily by physical structures. Related qualities, such as the authoritythat a dancer projects, are probably linked to personality. Rudolf Nuryey's dancing wascharacterized by spectacular leaps, strength, and intensity (Fonteyn, 1979).Mikhail Ba- ryishnikov's style is lighter, more androgynous, and more urbane. Differ- ent styles appeal to different audiences,but each audience is influenced by local and contemporary ideals of beauty. The Schools' Response to Dance Talent Even though several states include ability in the performing arts as part of their definition of giftedness, this partof the definition is virtually meaningless in all school systr -ns except the few that havespecialized schools in the arts, for example, North Carolina's School of the Arts or New York City's High School of Performing Arts. In most school systemsthere are no efforts to identifystudents who are gifted dancers, and there are no public school programs to develop talent in dancing. Thedevelopment of dance in this country is almost exclusively left to private danceschools, a circumstance which has several disadvantages, not the least of which is that much of the instruction in private dance schoolsis of poor quality (Braunschweig, 1981).

klentrwation.Gifted dancers are rarely identified through the public school system. Most pursue private lessons until they are outof high school and then compete for a place in musical theater,professional dance companies, or university dance programs. Their talent isevaluated through auditions. Dancers without formal training cannot compete suc- cessfully for a place in a professional ballet company, and wouldrarely be able to compete with dancers trained in jazz or modern dance. Aswith any

! 5 Children WA Talent in the Visual and Perforsehtg Arts 207 discipline, students with a history of good instruction are most likely to be identified as talented. In the United States, the population with any history of formal dance instruction is primarily composed of middle-class and upper-class females. Since most dance students train in private schools, the cost of instruction usually limits its accessibility to students from affluent families. Be- cause of sex-role stereotypes. most dance students are girls. The chances of an economically disadvantaged boys being identified as a talented dancer are very small. (See Box 7-1.)

BOX 7-1Break Dancing

Break dancing, a popular form of dance that received widespread publicity during the late 197Cisandearly 1980s, aptly illustrates the circumstances of economically disadvantaged males with dance talent. Although this group would not usually be identified as gifted in dancing, some talented minority adolescents were spotlighted for a short time. Break dancing belongs socompletely topoor, urban, black youths that it is, in a sense, an ethnic dance. It originated in the predominantly black ghetto of the South Bronx, N.Y. (Rosenwald, 1984). Until it gained attention nationwide, it was performed almost exclusively by members of the population that developed the dance. Break dancing embodies daring, physical prowess, and defiance. The dancers take turns executing difficult feats in an effort to out-do each other. The dance steps combine elements of acrobatics, the martial arts, and rock and roll dance movements. It is not a coincidence that break dancing was invented and performed by poor, black males, who, through the dance, set up and engaged in a competition that they had a chance of winning. (They also had a chance to pick up a little money from passersby who stopped to watch.) Break dancing's spectacular acrobatics drew notice from the media; and break dancers appeared briefly in movies, on television, and on concert stages. The;r popularity, however, was short-lived because the appeal of break dancing was primarily ott, novelty. Ironically, its domestication for the stage, television, and movies doomed break dancing to a short life on the streets as well. The middle-class's cooptation of a subculture's music, dress, or slang destroys its ethnic validity. The influence of cooptation is present throughout the history of black music, from early blues to modem jazz (Jones, 1971). Although the break dancing fad is an interesting social phenomenon for many reasons, it is of special interest to educators of the gifted in that it is an example of the limited forms of expression available to talented but disadvantaged youngsters. There were many dancers among this population whose talent became evident even though they had no formal instruction whatsoever. The situation illi.strates the point that talent alone cannot secure a dance career.

Programs.Serious dance training is rigorous; it requires it least an hour and a half a day at the dance studio (Braunschweig, 1981). For most talented dance students this study is not a part of their school curriculum. Dance in the elementary schools, when it is offered at all, is offerer{ in the

2 1 208 Ckildros with Taint t s Visual and Petforseing Arts form of folk or social dance in the physical education program. Even these programs do not often form a central part of thecurriculum, but are taught only if the physical education teacher happens to view them as important. Although learning creative movement skills, performing structured dances, and developing individual dances are considered to be important e'ements of early dance education (Kraus, 1969), theseskills are seldom taught. In the secondary schools, dance is also neglected. Where dance courses exist, they are often sustained only by aparticular teacher's knowledge and interest. One such program is the Marmaroneck (N.Y.) High School program, which has been in existence since the early 1970s. It offers four levels of dance classes and a planned curriculum (Barylick, 1983). The teacher of the program attributes its success to three circumstances: (1) availability of a dance studio (not a gymnasium to be shared with the physical education program); (2) a performing arts program including several teachers who act as advocates for each other; and (3) a curriculum that recognizes the value of dance as an end in itself, not as a means to developing spiritual values or staying fit (Barylick, 1983). A small number of specialized arts schools offer programs in dance on both a public and tuition-paying basis,depending usually on the student's place of residence. Some examples of' such schools are Houston's High School for Performing and Visual Arts (G. Jes 10-12), Cincinnati's Center for the Creative Arts (Grades 9-12); New York's High School of Performing Arts (Grades 9-12); the National Academy of Dance (Grades 7-13) at Champaign, Illinois; and the Philadelphia College of thePerfor- ming Arts (Grades 1-12) (Jacob, 1981). These schools usually select students on the basis of auditions and interviews and provide programsde- signed to meet both regular academic requirements and the needs of developing artists. Aside from these specialized schools, publicly funded dance programs are available almost exclusively atthe college level. This is too late to start the instruction of talented dancers.Moreover, most universities cannot provide the exposure to expert choreographersand dancers that association with a professional dance company can (Jacob, 1981). Choosing to purse a dance major in college is more common amongand more relevant to modernas opposed to classical orjazzdancers Uacob, 1981).

Acceleration and enrichment.Because there are so few sequential programs in the elementary and secondaryschools, there is little opportunity for any progress, much less acceleration. Even enrichmentefforts are minimal in most schools. An annualperforn ance of "The Nutcracker" constitutes the dance enrichment pi ,gram for many public school students. The ambivalence the public feels toward all the arts is less evident in 217 Children with Talent is the Visual and Perforweing Arts 209 their attitude toward dancethey are, apparently, fairly certain that dance has no place in the schools.

Acting and Actors Like dance, acting has a long history. In fact, it is only in the last 600 years that acting and dance have become twodistinct theater arts (Sorrel, 1967). The combination of dance, mime, song, and speech to enact dramas has been chronicled as early as 2,000 B.c. in Egypt (Kraus, 1969). The early Greek playwrights, including Thespis, were called "dancers," and Aeschy- lus instructed his own choruses in dancing and choreographed figures for them (Lawler, 1964). The primary medium of expression in acting is not movement, however, but words. Although gestureand movement are quite important in acting, the basic skills in acting are substantially different from those required for dancing.

Personality characteristics of actors.There are few studies of talented young actors, but at least one (Barron & Denman, 1972) reports that student actors, like other artists, are highly interested in their art and committed to perfecting their talent. These students also report that, since childhood, they have enjoyed doing imitations of other people's speech and mannerisms and have often impersonated others for the fun of it (Barron & Denman, 1972). Appar- ently many actors and comedians have entertained themselves through mimicry and the development of different personae long before they entertained others with their talent (Gardner, 1973; Lane, 1960). Imitating other people and enacting imaginary characters seem, in many cases, to be the potential actor's or comedian's way of coping with difficult situations. (Gardner (1973) specifically mentions school as one of the difficult situations that comedians have made more bearable for themselves by employing their talents.)

Cognitive skills in acting.According to the nineteenth century En- glish actress, Ellen Terry, acting requires intelligence, industry, and, most of all, imagination (Albright, 1967). More specific skills have been mentioned by others: the ability to discern and to imitate speech patterns and gestural mannerisms (Gardner, 1973), superior ability at comprehending verbal material (Wolf, 1981), and the ability to recall and to recreate emotional states (Stanislavski, 1948).

The Schools' Response to Acting Taleut Its close association with literature, an unquestionably academic subject, assures drama some attention in the public schools. The study of plays in English classes and occasional oral interpretation of literary selections 218 se.".

411011 Chi !dm with Taint is the Vinod aid Perfinsig Arts may afford some students very limited experiencein acting even where there are no drama classes. Experiences in drama are more readily available in the public schools than experiences in dan,.e. There are teachers who have had some training in speech and drama in most publicschool systems.

Identification if acting talent.In most school .wstems, identification of acting talent is not undertaken on a systematic basis. Because they are allowed some opportunities to act in elementary and secondary schooL however, a few children may be recognized as talented actors. Since most school systems offer no drama programs, talented young actors have to find private instruction to develop their talent. The young drama student is, however, at a much greater disadvantage than the talented young dance student, because dancing schools can be found even in small towns. Acting schools are, on the contrary, uncommon even in medit.m-sized cities. Chil- dren's theater is an option available outside of crhf.-1 .gn some communities, but it is only an introduction to theater. In the few public school programs with established identification procedures, selection is usually based on the student's commitment and superi- or acting performance asdemonstrated through audition. Preliminary auditions are sometimes submitted on video tapes and live auditions reserved for those students selected through review of the taped auditions.

Programs. The acting programs that do exist usually take the form of minkourses in enrichment programs at the elementary school level and electives, such as drama or oral interpretation, in the secondary grades. A few gifted programs specialize in the theater arts. For example, one federally funded prqiect for gifted disadvantaged students in a largely black and Hispanic urban school provided instruction in dance,music, drama, and graphic arts (Wolf, 1981). Most of the performing arts schools, such as those mentioned in the section on gifted dancers, also offer instruction in drama.

Acceleration and enrichment.In most public schools, acting is not taught on a formal basis, and the activities provided are not sufficient to foster progress in talent development. Typically, acting is offered as enrichment: Perhaps a few students are involved in production of occasional plays for the student body, a class of students may attend a performance at a local theater, or a few students mayparticipate in interpretive reading competitions. Like dance training, dramatic training is not available to most students until they enter college. Performing arts schools are the only public school programs in which the curriculum is compreher and sequential enough to allow acceleration.

2 I Children with Talent in the Pima! asd Perform* Arts 211

SUMMARY

Artistic talent is distinct from creativity, and it cannot be assessed using the instruments developed to identify creativity. For a variety of reasons, artistic talent is better viewed as a number of separate talents. In most cases, the reseaych reported here noted that above-average intelligence seemed necessary for a high level of talent development. In arts other than the visual arts, performance ability differs in some ways from the ability to make works of art. In dance and music, however, the chapter reported that interpretation was closely related to the ability to make works. In music, particularly, excellent musical performance seems to have a great deal to do with the musical understanding necessary for composition. Most artistically talented students share an uncommon degree of devotion to their work. Whether they study dance, acting, music, or visual arts, the intellectual interests of talented young arts students seem in all cases to extend to issues of aesthetics and philosophy. Research also confirms that arts students are critical nonconformists. During their school years they feel isolated and out of touch with their peers and with their experiences in school. Though little support exists for public school education in the arts, the research discussed here strongly suggests a determining role for early nurture and training in the arts. Early, high-quality instruction seems to be crucially important for musicians and dancers, and itis apparently an influential experience for visual artists and for actors as well.

220 '!PL

Typical Gifted Development

I.Focusing Questions h.Types and Stereotypes A.Nineteenth Century Views B. Terman's Findings 1. Physical development 2. Emotional development 3.Intellectual development 4.Problems with Terman's findings C.Recent Studies of Preschool Gifted Children 1. Developmental patterns 2.Importance of early environment 3. Prenatal and perinatal environment 4.Accelerated early teaming 5.Problems of highly gifted young children M.Developmental Domains: Types and Anomalies IV.Language Acquisition and Verbal Precocity A.Theories of Language Acquisition 1. The nativist theory 2.The behavioral theory of language learning B. Precocious Language Acquisition and IQ 1. Recent research C.Verbal Ability and IQ 1. Lack of instruction as a confourNiing variable D.Early Reading 1. Characteristics of early readers 2.Other abilities and interests of early readers

212 221 Typical Gifted Developnera 213

3.Families of early readers 4.Perceptual skills in early reading 5. Reading achievement and decoding speed 6.Gifted children with reading problems V.Concept Development A.Piagetian Notions: Stages and Adaptation 1.Sensorimotor stage 2.Preoperational stage 3.Concrete operations 4.Formal operations 5.Adaptation B. Piaget and the Gifted 1. Gifted children and stage attainment 2.Mmit issues 1Implications for concept development 4.Piaget and learning style 5. Cautions VI.Social and Emotional Development of Gifted Children A.Gifted Children as "Supernormal' 1. Social interactions 2.Emotional stability B. Gifted Children as Deviant 1. Effects of deviance C.Emotional Problems, Real and Imagined D.Real Problems of Gifted Children 1. Peer relationships 2.Oversensitiveness E. Incidence of Severe Social and Emotional Problems 1. Delinquency 2. Mental disturbance 3. Suicide VII.Summary

Foaming Questions 1. In what ways do Lombroso's and Terman's views of gifted individuals represent conventional stereotypes? 2. How does the early devefopment of the oral language of gifted children compare with that of other children? 3.To what degree do intellectual skills, perceptual skills, and environmental support correlate with early reading ability? 4.What do Piagetian measures suggest about gifted children's concept development? 5.What factors contribute to the popular perception that gifted children are at risk for social and emotional problems? 6. How does most recent research characterize gifted children's social adjustment and emotional health?

22 2 214 Ivied Gifted Development

TYPES AND STEREOTYPES Considering the fact that gifted children are identified because of their atypical development, a discussion of the typical development of gifted children may seem problematic. A number of writers, however, have tried to describe characteristic patterns of development among gifted children. A review of the ideas presented by the most notable of these writers provides a basis for distinguishing between accurate representations andstereotypi- cal views of gifted children's development. Nineteenth Century Views The ideas developed in Darwin's The Origin of Species, published in 1859, prompted a number of nineteenth-century thinkers to apply the notion of natural selection to the study of human characteristics. Intelli- gence (genius and feeble-mindedness) was among thecharacteristics most often explained in term- of Darwin's theory. The reasoning behind this application of Darwin's theory was as follows:

I.The characteristics of all living things are hereditary. 2.The most successful characteristics are those that provide the individual with an adaptive advantage. 3.Those individuals who adapt better have a greater chance of producing offspring. 4. A greater amount of intelligence is an adaptive advantage. 5.Hence, intelligence is hereditary and the most intelligent members of society should be the breeders of offspring (and less intelligent members of society should be kept from breeding).

Based on these premises, the nineteenth-century hereditarians began to study the traits of individuals who werefeeble-minded and those who were eminent. Galton (see Box 8-1), for example,studied eminent families in several occupational categories. He concluded from his study of these families that "genius" was an inherited trait of positive benefit to society (see Chapter 4). Lombroso, by contrast, concluded from his study of eminent men that genius and degeneracy were closely connected. He characterized the genius as a sickly and unstable type who was likely to be antisocial and to die early. Lombroso saw normality as the characteristic of most unqualified benefit to society. He was suspicious of any form of abnormality, including abnormal brilliance. Although these nineteenth-century views of genius differed greatly, neither was derived from what we now consider to be a scientific sampling of highly intelligent or creative individuals. The work of these investigators raised the issue of extreme talent more extensively than ever before, but their work also tended to substantiate unfounded prejudices.

223 Typical Gifted Develop:sew 215

BOX 8-1Sir Fratwis Gallon as a Gifted Child

Sir Francis Galton, who is noted for his statistical study of eminence, washimself a highly gifted child. According to Tennan, Gallon probably had an IQ of 200 (Burt, 1975). The last of nine children, Francis received his early education at home withhis sister Adele serving as his first teacher. Francis' precocious developmentand early tutelage resulted in highly advanced academic development. By the age of one year, Francis could recognize letters. By the age of three, he could read children's stories and by four could read adult literature. At six, Galton was able to read the classics in both Latin and Greek and had reportedly memorized the Odyssey and the Iliad. When he was eight, Calton was sent to a private school where he was placed in aclass with fourteen- and fifteen-year-old boys. Here, and in his later placement at King Edward's School in Birmingham, Calton found the pace of the curriculum slow and the content boring. His interests in science and mathematics were notsatisfied by the classical education he received at these schools. Considering his own experience as a highly precocious youngster, it is not difficult to understand why Galion spent a large part of his adult life speculating about genius. Although his views about the nature and transmission of intelligence have been dis- puted and discredited, Galton's statistical methods for studying human characteristics served as the basis for later studies in the nineteenth and twentieth centuries.

Terman's Findings In 1921 Lewis Terman began a longitudinal study of 1,528 high-IQ children in California. He was interested in identifying characteristics common to gifted children and to tracethe life histories of such children (Seagoe, 1975). As a result of this study, Terman and his associates concluded that gifted children were intellectually, physically, and emotionally superior to average children. These results tended to refute the beliefs of nineteenth century theorists like Lombroso.

Physical development.Although the physical differences between gifted children and average children were not dramatic, they were statistically significant.Termanfound the gifted to be taller and heavier at birth, to walk and talk earlier, to start puberty earlier,and to be healthier than average children. Rather than being prone tophysical weakness aswas previously thought, gifted children appeared to be athletic, muscular, and well-nourished.

Emotional development.Terman fbund that gifted children sur- passed average children in emotional and moral qualities.Although he acknowledged that the tests of character he used were not highly reliable, Terman nevertheless concluded that the gifted group was superiorin character to the control group. He foundgifted children to demonstrate creativity, aesthetic sensitivity, perseverance, self-confidence,cheerfulness, achievement motivation, a sense of humor, and leadership ability. He 214 216 Typical Wird Depsiapsuia found them to be more emotionally stable and mot . adaptable socially than average children.

Intellectual development.Children with IQs above 140 were selected for Terman's study. The average IQ of the group was 151, and the range of IQs was from 140 to 200 (Seagoe, 1975). Although the children in Terman's sample represented the top 1 percent on measures of intelligence, they were not quite so superior inacademic achievement; their average achievement quotient was 144. Only half of thechildren in the sample had learned to read before entering school, and on average the gifted children tended to be performing school work at a level just two years above their chronological age expectancylevels (Burt, 1975). Terman was disturbed by the fact that children who werehighly superior in intelligence and quite superior in academic achievement, wereonly slightly advanced in their grade placement in school (Laycock, 1979).

Problems with Terman's findings.Bias in the selection of his subjects produced significant problems with Terman's findings. Because most of the children Terman tested were those whom teachers nominated, he failed to testand thereby identify as giftedmany other equally bright children. Considering the usual biases in teacher referrals, these unidenti- fied gifted children may have been underachievers or troublemakers. They were likely to have come from lower class orminority backgrounds. Because children from middle-class and upper middle-class backgrounds were overrepresented in Terman's gifted group, some of the findings about gifted children might in truth have been findings about children from advantaged backgrounds. The greater-than-average birth weight of the gifted children in the study, for example, may have been a phenomenon more stmngly associated with theirsocioeconomic status than with their IQ. In considering the typical development of the gifted, we might be just as incorrect in accepting Terman'sfindings about their normality as we are in accepting Lombroso's speculations about their deviance. The obvious empirical advantage of Terman's methods over those used by Lombroso and Galton may still not have provided an accurate picture of the developmental history typical of gifted children.

Recent Studies of Preschool Gifted Children Some researchers (e,g., Roedell et al., 1980) have taken aspecial interest in the very early development ofgifted children. Usually, such researchers discuss the developmental histories of highlygifted children since these children appear unusual almost from birth.Young gifted children of less extraordinary capability are more difficult toidentify: they

225 rypiceal Wird Develop:mat 217 may show some degree of precocity in one or moreof the developmental domains, but their giftedness may not be apparent until they areolder.

Developmental patterns.Retrospective studies (Hollingworth, 1942; Terman, 1925) of children with very high 1Qs have noted their precocious verbal development. By the time they are two or three years ol I, these children demonstrate a speaking vocabulary that is larger than averageand sentence production that is more advanced.With parental encouragement and assistance, they are often reading a little and doing someaddition and subtraction before they start kindergarten. These high-IQ children also show early and rapid development of conceptual skills such asdiscriminat- ing and labeling colors, drawing, identifying shapes, and workingpuzzles. In the development of motor skills, preschool giftedchildren are most like other children. On tests of grossand fine motor skills, preschool gifted children may score slightly higher than average, ...ut not nearly so high as they do on cognitive tests (Kitano & Kirby, 1986; Roedell etal., 1980).

Importance of early environment.Recent studies of animal behavior suggest that learning can be greatly acceleratedby early stimulation (Clark, 1988). In fact, we might reasonably speculate that, given appropriateearly stimulation, most children cou?d develop cognitive abilities in the gifted range. Even physical development,which seems most influenced by chronological age, can be accelerated considerably. Stimulation of early learning seems to depend on a responsive environment. Simply to surround a baby with pictures, sounds, andobjects may not be enough,however. The baby also needs to interact with the environment, and his or her actions must have some effect.Learning takes place when the baby's actions have predictable consequences. Parentswho respond to their infant's crying, gurgles, and gestures give theirbaby an opportunity for this kind of learning. Such parents are morelikely than less responsive parents to have high-achieving children.According to Clark (1988), every child needs r) feel effective in order tolearn at an optimal rate.

Prenatal and perinatal environment.Although a poor prenatal environment is detrimental to babies, a good prenatalenvironment may not be exceptionally advantageous. It is difficult to demonstratethat excellent prenatal conditions contribute to the development ofsuperior intellectual capacity. Since bad environments inhibit development,however, we might infer that good environments may stimulate it. We can make the same sort of inference about theconditions of birth. H' serious birth trauma, such as deprivation of oxygen, canimpair intelligence, then unusually goodcircumstances at birth may enhance intelli-

226 218 nokd Gifted Deve lopsind gence. Babies of unmedicated mothers are more alert immediately after they are born, although it is not certain that the effectsare long lasting. Clark (1988) reports a study that suggests that increasedoxygen to the brain can 6:Tease the intelligence of infants. Shenotes, however, that there is insufficient evidence to be certain of these findings.

Aceekrated early learning.Recently, some authors have suggested that children's development can be accelerated significantly by providing them with a program of highly stimulating experiences (seee.g., Prichard & Taylor, 1980). One of the most outspoken proponents of early stimulation is Glenn Doman, founder of the Institute for the Achievement of Human Potential 21d author of How to Teach Your Baby to Read (1964). His training program purports to teach children how to read atone year of age, how to do simple arithmetic at the same age, and how to speak a foreign language. Doman contends that, given an intensive program of enrichment, any normal child can learn skills at an accelerated rate. How- ever, his claims are supported only by anecdotal evidence and not by controlled studies. The testimonies of parents who have participated in his program are mixed (Langway, 1983). Some professionals feel that environments of the sort Doman advocates are too stressful for young children. These psychologists and educators believe that accelerated early learning causes children to have social or emotional problems, and perhaps even academic ones. There is, however, little evidence to support these assertions.

Problems of highly gifted yotmg children. One of the major problems of young gifted children is the disparity between their intellectual development and their physical development. As Roedell, Jackson, and Robinson (1980) point out, it is hard for teachers to accept that three-, four-,or five- year-olds whose writing is as laborious as that of their classmates can read and perform arithmetic at advanced levels. Sometimes teachers are unable to recognize children's precocious cognitive development because they focus on the children's average motor performance. Kindergarten teachers, in particular, have trouble identifying the highly intelligent children in their classrooms; theymay fail to recognize the ability of 90 percent of their gifted students ( Jacobs, 1971). Even when they are identified, such children may not receive appropriate educational programming because it is difficult for schools to accommodate the disparity between young gifted children's skills in comprehension and their skills in written expression. Young gifted children may also experience difficulty because of their advanced social development. Parents may not be able to find suitable playmates for gifted children whose interests and patterns of play are quite

227 rypkal Gifted Developsevat 219 different from those of peers. Some authors (e.g., Freeman, 1979) point out that gifted children select older children asfriends. Highly gifted children may also prefer the company of adults (Austin & Draper, 1981).

DEVELOPMENTAL DOMAINS: TYPES AND ANOMALIES

Now that we have considered several general perspectives on thedevelopment of gifted children, we can look morespecifically at the empirical research. In the subsequent part of the chapter we will review the research on the development ofgifted children in three domains: language acquisition, concept formation, and social and emotional development. The findings discussed in these sections do not reflect a unified perspective on the development of gifted children, however. UnlikeGalton, Lombroso, or Terman, we are not attempting to present a typology of giftedness. Nor are we attempting to characterize the early developmentof highly gifted children. Instead, we examine the development ofparticular skills in gifted children and contrast the rate and sequence ofgifted children's development with that of average children. It is important to remember that the empirical findingsabout gifted children's development merely reflect trends. These findingsdo not account for the individual variations indevelopment that occur among the gifted. Many gifted children do not conform to the typical sequenceof development; others conform in some domains but not in others. To say, for example, that gifted children areemotionally more mature than average children is not to saythat every gifted child is more mature emotionally than every average child. In particular, the findingsabout typical gifted children may not describe the actual experiences ofunderachieving children, disadvantaged children, or highly precocious children. The developmental patterns characteristic of these types of gifted childrenwill be discussed in later chapters.

LANGUAGE ACQUISITION AND VERBAL PRECOCITY

Because language, written and oral, is the medium of formalinstruction and of formal learning, its acquisition and early developmentseemin some unclear wayto prefigureschool learning. Are precocious talkers likely to become high-IQ schoolchildren? Do gifted children tend tobe early readers?

228 220 Typical Gifted Divelopand Theories of Language Acquisition Some quite plausible theories have been developed to explain the way in which language is acquired by young children. Recent research draws on these theories to help explain the typical development of gifted children.

The nativist theory.Modern linguists have constructed the "nativist" or "innatist" view of language acquisition. Just as some theories of intelligence maintain that intellectual ability is an inborn trait (see Chapter 3). so the nativist theory of language acquisition maintains that language ability is an inherent trait of human beings. There are, however, some major differences betueen nativist views of language acquisition and hereditarian views of intelligence. Whereas theories of innate intelligence tend to stress individual differences in intellectual ability, the nativist pusition on language acquisition tends to stress the commonalities of language ability in human beings. The ability to use language fluendy is considered by theorists like Noam Chomsky to be the genetic heritage of all human beings. The nativist theory credits environmental influences as the source of the vast cultural and individual differences evident in language performance. It credits genetic andbiologicalinfluences as the source of a common competence. The essence of the nativist view of language acquisition is the image of children as langnage rule makers. A child who says, "Mammy goed away" instead of "Mommy went away" demonstrates this rule-making ability. Al- though the child has probably heard the form "went" countless times, the internalized logic of grammar leads the child to say "goed". In a logical sense, the child's mistake is not a grammatical error.

The behavioral theory of language learning.According to this view, language is learned through the mechanism of operant conditioning, and it is subsequently used by the learner to adapt the environment. Behaviorists view language as a set of stimuli and responses. The internal mechanisms, the cognitive apparatus, and the developmental stages that may regulate language acquisition in young children are of little interest in the behavioral view. The practical interactions that condition specific language uses are, however, of immense interest to behaviorists. Linguistic meaning and structure are of less interest in this view than the way in which verbal units condition behavior. The behavioral and practical content of language, not its phonology, morphology, or syntax, is of most interest to behaviorists.

2 Typical coca Devetkpnerd 221

Precocious Language Acquisition arKI IQ Early speech is almost universally accepted as a token of a child's general aptitude for the important tasks of reading and writing. Fond parents are apt to look for signs of verbal precocity in their infants, and to find evidence for it in early speech. Hollingworth (1942) reported the onset of speech in the case of 12 children with IQs above 180. Although the children are described as early talkers (Hollingworth, 1942), the data, supplied by parents, do not support that conclusion. The youngest age at which any of the children began talking is reported to be 9 months and the latest age is 24 months. The median was 14 months, approximately the same as for average children.

Recent research.Recent studies that draw on both theories of language acquisition (often in the same study) may offer someinsights into the language development of gifted children. Smolak (1982) studied cognitive precedents of receptive and expressive language in infants and beginning talkers (n = 48) tested at 9, 11, 12, and 15 months. Smelak reported that infants whose parents played verbal games with them formed better on comprehension tasks, and that children who perfs, . zd better on comprehension tasks were more likely to talk. Evidently, being talked to is strongly related to talking, a finding that is not surprising. Guilford, Scheuerle, and Shonburn (1981) studied language acquisition in a small sample (n = 11) of preschool children with Binet IQs above 140. These researchers concluded that their subjects did not have a better command of grammar than average children. They concluded, however, that their young subjects could apply more fully than normal subjects the rules that they knew. This study implies that it may be diffic:31t to identify young gifted children on the evidence of their oralexpressive language. Gray, Saski, McEntire, and Larsen (1980) to similar conclusions in their study of oral proficiency as a predictor of academic success. They reported that when the effect of IQwas controlled, "little relationship is evident between oral language proficiency and school readiness" (Gray et al., 1980, p. 266). At present, research does not suggest that early language development, apart from IQ, is a very reliable predictor of' later academic success. It is possible that this conclusion may be the result of the way in which early language is studied. Most studies involve small samples in a cross-sectional design. This topic of research is sure to remain of interest, however, because of the apparent conflict between theories of language that stress the important influence of environment and theories ofintelligence that stress the important influence of genetic inheritance. 112 7)pical Gifted Divelopainst Verbal Ability and IQ For some time the literature on gifted children has suggested that these children possess advanced verbal abilities (cf. Terman & Oden, 1947). It is possible to specify the confirmed verbal skills that characterize gifted children. These major characteristics, which involvereceptiveverbal skills almost exclusively (cf. Gallagher, 1985), are as follows:

They have vocabularies significantly larger than average children. They tend to read earlier, to read better, and to read more than average children. They comprehend difficult verbal material more easily than average children.

Receptive verbal tasks are, in fact, those most frequently sampled on IQ tests like the Stanford-Binet (Terman & Merrill, 1973), the Wechsler Verbal Scales (Wechsler, 1974,1981), and other such tests. Since the development of receptive skills probably precedes the development of expressive skills, this method of assessing academicpotentialmakes a good deal of sense.

Lack ofinstruction as a confounding variabk. When the expressive verbal performance of gifted children has been evaluated, as in the studies of language acquisition reported previously or in studies of the writing of older students (e.g.. Fearn, 1981; Pendarvis, 1983), results have generally confirmed the similarity of gifted and average students. Why should such a discrepancy exist between receptive and expressive language performance in the case of gifted children? The fault, according to William Durden, who directs the Program for Verbally Gifted Youth (PVGV) at Johns Hopkins University, seems to lie in students' instruction, not in their academic potential. He cites the neglect of the humanities and written expression in elementary and secondary schools as a probable cause for the difference between gifted students' receptive and expressive language skills (Durden, 1980).

Early Reading Although it is uncommon for children to be taught to read before they enter first grade. a small minority do learn to read as preschoolers. Of these early readers, most start reading at four or five years of age; a few begin even earlier. Early reading is an important issue for educators of the gifted because many, but by no means all, gifted children are also early readers. The following discussion reviews the research related to early reading by examining several pertinent questions: 231 Typical Gifted Developotesst 223

I.Do early readers share cognitive characteristics? 2.Does early reading correlate with high IQ? 3. Do early readers come from homes that encourage early reading in particular ways? 4.Does early reading relate to petteptual skill development?

Characteristics of early readers.Two general patterns emerge from studies of early readers: (1) these children have above-average intelligence and (2) they have been exposed to written language and other academically pertinent modes of expression at home. The parents of early readers support their children's efforts to read, even if they do not provide direct instruction. In addition, there are some specific interests and abilities common to many early readers and their families. Evidence of the importance of intellectual precocity in learning to read early is found in many studies. One of the oldest (Davidson, 1931) is described in Durkin's (1936) Early Readers. Davidson selected 13 children ranging in chronological age from three to five years old, but having in common a mental age of four years. After receiving ten minutes of daily reading instruction for four and a half months, the children were tested. All of the children learned to read some words, but the bright three-) ear- old children learned more than the older children, a finding that suggests that ever. moderate intellectual precocity influences early reading achievement. Davidson's study also suggests that children can learn to read considerably earlier than the age at which they are typically instructed. Estimates of the proportion of early readers among high-1Q children range from one-third to one-half. Terman (1925) surveyed the parents of 643 children with IQ scores of 140 or above to determine the incidence of early reading. He found that approximately 45 percent of the children were reported to have read before starting school. Another survey based on self-reports by gifted junior high-school students, found that about half had learned to read before entering first grade (Strang, 1954). Among highly gifted children, the incidence of early reading seems to be greater. All of the 12 children described by Hollingworth (1942) in Children with Above 180 IQ were early readers. Of 270 seventh and eighth graders who made composite scores of 1,100 or higher on the Scholastic Aptitude Test (Verbal and Math), 81 percent read before they were six years old (Van Tassel-Baska, 1983). Fifty-five percent of the group read before they were five years old. Results of some studies of early reading, however, must be interpreted cautiously. Studies that rely on parent reporting or self-reporing may not be very accurate. Durkin (1966) avoided themethodological problem of self- or parent reporting. By testing 5,103 first graders in Oakland, California elementary schools during the first weeks of school before read-

2 3 2 224 Txpied Gifted Devaleponst ing instruction had begun, Durkin and her co-workers were able to identify 49 (approximately 1 percent) who could lead. Criteria for selection were (I) correct pronunciation of at least 18 of 37 vocabulary words and (2) a raw score of at least 1 on a standwdized reading test. The children's reading grade levels ranged from 1.5 to 4.5. The IQ range in the Oakland study was 91 to 161, with a mean of 122. The correlation between reading achievement and IQ was 0.40 In a later study in New York City, employing similar identification methods, Durkin found 156 early readers among 4,465 children. Their reading grade levels ranged from 1.4 to 5.2. The IQ range was from 82 to 170, with a mean of 133. The correlation between reading achievement and intelligence in that study was only 0.24. From such findings it is clear that not all early readers are high-IQ children; neither are all high-IQ children early readers. Durkin's low to moderate correlations might, however, reflect the effect of "restricted range." This concept refers to the fact that when the range of one of two correlated traits is "restricted" (i.e., reduced), their correlation will be reduced. Durkin's students exhibited both elevated IQ and elevated reading achievement. In the New York study, for example, 78.2 percent of the early readers had IQ scores of 120 or above. In the general population the correlation of early reading and IQ may well be higher than Durkin's reported statistics. Correlations of early reading skill with various intellectual measures in the range of 0.30 to 0.64 are nonetheless typical for this age group (Hiebert. 1980; Stanovich et al., 1984). Correlations of this magnitude suggest the importance of identifying additional factors that contribute to precocious reading achievement.

Other abilities and interests of early readers.Durkin described the early readers in her study as "paper and pencil kids" and as "scribblers." Their interest in reading, in fact, seemed often to derive from an interest in copying words and letters. An early interest in letters of the alphabet and in words is also reported in other studies of children who read early (e.g., Hollingworth, 1942; Salzer, 1984). Of course, not all gifted children share this important characteristic of' early readers. Another characteristic of early readers seems to be their interest in sedentary games and activities. The parents of early readers described their children as being adept at games and activities such as drawing (Dur- kin, 1966). Parents of equally bright children who were not early readers did not ascribe this characteristic to their children. Other than an unusual interest in letters and words, little has been found to distinguish early readers except a difference in their families' tendency to encourage reading.

233 Typical Gifted Devekinsest 225 Families of early readers.Reading early is accompanied by some form of assistance. Even when the parents of early readers are unaware that their children are learning how to read, their reports suggest that they provide an environment that supports early learning efforts. Compared with the mothers of equally bright children who do not read early, the mothers of early readers spend more time interacting with their children in educational games and in the daily routine of housework, meals, and recreation (Durkin, 1966). They read aloud more to their children, and spend more time providing assistance when their children show an interest in learning how to read. Parents of early readers are also more likely than other parents to describe themselves as avid readers. At the time that Durkin's studies were conducted, many parents were reluctant to teach their children how to read because of educators' admoni- tions that to do so would be detrimental to the child's academic adjustment (Durkin, 1966). The parents of early readers tended to be unconvinced by these warnings (Durkin, 1966). Hiebert (1980) found, however, that home experiences (including parental instruction) could not alone account for young children's print awareness. Parental teaching, in combination with high IQ and advanced Piagetian cognitive development did account for 59 percent of the variance (coefficient of multiple correlation = 0.77) in children's print awareness. (See a later section in this chapter for a discussion of Piagetian cognitive development.)

Perceptual skills in early reading.Learning how to read may make heavier demands on perceptual processes than on cognitive development. Abilities associated with perceptual development may therefore also help explain individual differences in young children's reading. This view is supported by studies that have found a lower correlation between IQ and reading achievement in earlier grades than in 1er grades (Stanovich et al., 1984; cf. Carter & Kontos, 1982). After child. ster basic phonics rules in the early grades, they begin to encounter se- mantically and syntactically complex material. Then the demancis 01 the reading task seem to change dramatically (cf. Chall, 1967).

Reading achievement and decoding speed.Decoding speed is the speed with which readers can translate visual symbols into meaningful auditory signals. Research offers evidence for the importance of decoding speed in reading (e.g., Hunt, 1978; LaBerge & Samuels, 1974; Lesgold & Perfetti, 1978; Stanovich et al., 1984). Taking issue with Arthur Jensen's (1980) assertion that reading achievement is largely a function of general intelligence, Stanovich, Cun- ningham, and Feeman (1984) studied the relationships between first

234 226 rifledGijkd Dnolaftwa graders' reading performance and their phonological awareness, decoding speed, listening comprehension, and general intelligence. The strongest correlation with reading achievement was decoding time (-0.52). Stanovich and associates (1984) reported that the comparative importance of decoding speed versus IQ clearly emerged in multiple-regression analysis. Their analysis suggested that pseudoword naming (i.e., the speed of decoding nonsense words) made the most significant independent contribution to predicting reading achievement in first-grade children. When combined with measures of phonological awareness, the addition of decoding speed accounted for substantially more variance in reading achievement than did the addition of intelligence measures. Most studies have shown a decrease in the importance of decoding skills at later grade levels (cf. Chall, 1967). It seems likely that as reading materials become more conceptually difficult and as readers become more fluent, then their perceptual skills (such as decoding speed) become less relevant to the task of reading.

Gifted children with reading problems.Students identified as gifted by virtue of their high scores on IQ tests can usually read fluently by the end of the first grade; a significant proportion begin to be fluent readers somewhat later; and a small percentage experience much difficulty in learning how to read. A few high-1Q children have persistent reading problems in school. Pringle (1970) found that about 25 percent of the underachieving gifted children in her study were reading below grade level. Whitmore (1980) also identified several primary-grade gifted students who read below grade level because of learning or behavior problems. Fox (1983) studied children with scores of 125 IQ or higher on the Wechsler Intelligence Scale for Children-Revised (WISC-R) who had been referred for testing because of their poor school performance. The study found that 75 percent of these students were reading at or below grade placement level and 10 percent were reading two or more grade levels below grade placement as measured by a standardized instrument. At the same time, Fox found that 89 percent had listening comprehension levels two or more years higher than their reading level, as measured by an informal reading inventory. It is clear that many preschool children with average or higher IQ scores are eager to read ant: have the ability to learn. The continued superior reading achievement of precocious readers (whether gifted or not) shows that to encourage and to support early reading does no harm to children's academic development (Durkin, 1966). For the relatively few gifted children who find early reading difficult, individualized instruction in reading has been shown to effect significant improvement.

2 :1 5 Typkai Gilied Development 1.27

CONCEPT DEVELOPMENT The study of concept development isperhaps the most theoretical pedagogical specialty. Its subject is how ideas emergeand mature during the growth of humans. Concept development is a field so broadthat it necessarily impinges on every branch' of psychologyand pedagogy (Flavell, 1977). It draws on studies of perception, memory,motivation, and social competence. In order to focus dixussionhere, however, this section of the chapter will consider primarily the work of the Swissscholar Jean Piaget. In the last 20 years Piagetian theory has guided thestudy of concept development in the United States andBritain. Piaget's lifework was an investigation of the way inwhich certain concepts emerge in the human being.Piaget believed that the concepts he chose to investigate were important categoriesof human cognition. These concepts appear in Piaget'swork as "pure" (i.e., nonverbal) concepts. Piaget built his theory of learning on a profoundunderstanding of epistemology (i.e., the philosophy of knowledge).This grounding sets Piaget apart from most U.S. and British researchersof cognitive development. Most research oncognitive development in the United States and Britain addresses the means by which individualslearn, not the interaction between the categories of knowledge and learning. Because the gifted are seen to be facile abstractthinkers and to enjoy the consideration of theoretical issues, itis important to examine their development of concepts. Before we review theempirical literature on Piagetian concept development among the gifted,however, a brief review of several of Piaget's most important ideas isin order. Piagetian Notions: Stages and Adaptation Piaget maintained that individuals progress throughdiscrete yet hierarchical stages of concept development. According tohis theory, each stage involves characteristic modes of thinking. Children progress attheir own rates from stage to stage by usingthe thinking modes of the preceding stage to propel them into the nexthigher stage. Piaget's concern is with the quality of thought at each stage, not with the rate atwhich individual children reach particular stages. Piaget also described the process by whichchildren uome to make the transition from stage to stage or to incorporate newinformation into their existing structure of thought (called a schema).This process of adaptation involves two mechanisms, assimilation andaccommodation.

Sensorintotor stage.The first stage takes infants from thelevel of thought that is totally constrained by immediateexperience to a level of thought that transcends immediate experience.This ultimate mental ac- 2 n 228 Typical Gifted Development complishment of the sensorimotor period is apparent in events such as babies' search for objects that are hidden and their crying for a parent who leaves the room. Even at the height of the sensorimotor stage, however, children's thinking is rooted in the concrete persons, things, and events of their immediate experience.

Preoperational stage.In the preoperational stage, children begin to develop symbolic patterns of thinking that transcend immediate things and events. Children develop concepts of size and quantity, akhough these concepts seem to depend on particular perceptual conditions. For example, preoperational children can distinguish between more and less on4 on the basis of one variable. They can tell whether there is more water in one glass than in another based only on the height of the water line. They cannot conceive that the width of the glass is a second variable that affects the amount of water in the glass.

Concrete operations.At this stage of development children are able to handle the abstract relationships among objects. They are able to understand that more than one variable can affect the properties of objects, and that objects can be manipulated without altering their essential properties. This latter process, Alen termed "conservation," allows children to understand, in concrete terms, mathematical concepts such as the commutative property of addition and multiplication, the principle of identity, and the reversibility of operations. Although children at this level can manipulate objects abstractly, they cannot yet generate and manipulate verbal ideas that symbolize the relationships among objects and events For instance, children at the stage of concrete operations can understand the principle of identity; however, they would be unable to use th.- principle of identity to manipulate variables in an algebraic equation.

Formaloperations. The greatest degree of abstraction of thought is possible at the formal operations stage. Children at this le.vel of development are able to derive conclusions from hypotheses thin are generated verbally. Their thinking is not constrained by the world of objects nor even by the rules that typically govern the relationships among objects in the rea: world. At this stage, children can speculate about hypothetical reality. They can generate logically consistent systems that need not be verified empirically.

Adaptation.Piaget equates mental development to the biological process of adaptation. Just as all living things behave in ways that allow them to adapt to their surroundings, so thinking beings adjust their methods of thought in ways that allow them to adapt to the symbolic events that 23 7 Typical Gifted Devekfatest 229 they encounter. Structural constraints limit biological adaptation, and the constraints of preexisting modes of thinking (schema) limit conceptual adaptation. According to Piaget, however, cognilive limits can be stretched through a process called "equilibration." True equilibration can only occur when the mind changes its schema in response to new information. This process is termed, "accommodation." Children move from one stage to the next higher stage through this process. Another way that a child can handle new information is through the process of "assimilation." This process allows the child to fit new information into an existing mental structure. A simplistic example can help illustraw these two processes. Let's assume that a young child (a girl in this exampk) has never seen a cat but has had plenty of experiences with dogs. When the child sees a cat for the first time, she is likely to call it a dog (that is, to think of it within the constraints of her existing mental schema). When the child calls the cat, "dog," she is assimilating the cat into her existing schema. If, however, the child distinguishes the cat from her experience of dogs, she may call the cat, "animal" or ask someone else what the strange animal is called. This latter process of accommodation allows the child to derive a new concept based on greater abstraction. In this case, the realization that the cat is not a dog has forced the child to derive the more abstract notion of animal. According to Piaget, such processes are not, however, dependent upon the language used to describe them. In this case, the child's language reveals the underlying mental process even though the mental process is not based on the linguistic act of naming the object.

Piaget and theGifted Because it has so often been claimed that gifted children are by nature better at understanding and using concepts and at thinking abstractly, the Piagetian studies that use gifted subjects are of particular interest. Gifted children (e.g., high-IQ children, precocious math and reading students) figure prominently in some of the empirical investigations based on Piaget's theories. The work of Piaget seems to present theoretical clues for the empirical study of concepts (Carter & Kontos, 1982; Keating, 1980b). A critical task of researchers has been to translate Piaget's largely philosophical insights into a manageable empirical framework. Many of the early investigations of Piaget's hypotheses were flawed attempts at such translation, while others were more limited attempts to validate one or another of Piaget's untested hypotheses. More recently, better empirical studies (e.g., Carter & Ormrod, 1982; Keating, 1975) have begun to appear.

Gyied children and stage attainment.The central tenet of Piaget's view of concept development is the progression of human beings through 230 7}Itical Gifted Develefesand an invariable sequence of four fundamental stages: sensorimotor, pre- operations, concrete operations, and formal operations. Children must, according to Piaget, progress through each stage in succession. Given this premise, a question of concern to American researchers has been whether or not early attainment of stages is characteristic of gifted children. The results of the research are not clear, however. Some studies (e.g., Moore, Nelson-Piercy, Abel, & Frye, 1984; Norton, 1980; Smolak, 1982) have yielded weak or insignificant correlations between attainment of the early stages and measures of cognitive abilities such as IQ, language production, and reading achievement. By contrast, the association of high IQ with the early attainment of forniel operations seems to be stronger (Car- ter & Kontos, 1982; Carter & Ormrod, 1982). Why should this be the case? A logical answer lies in the approximation of high-level performance on IQ tests to formal operations, as opposed to other stages. Formal operations pertains clearly to many of the symbolic notions of conceptual mathematics (e.g., the properties of equality and of operations) and to facility in the formal study of language and literature. Precocious children acquire these skills early, and they are also likely to perform well on IQ tests. Carter and Kontos (1982) claim to t*-ve found that gifted children achieve the stage of formal operations about two years earlier than children with IQs in the average range.

Moot issues.One problem has stymied American researchers for some time. Unless carefully controlled distinctions can be made within Piagetian stages, then the association of IQ and stage attainment or acceleration of stage progress cannot be measured with very great accuracy. The inaccuracy results from the fact that children are expected to remain at a particular stage for quite a long period of time (e.g., approximately four years at the concrete operations stage). Piaget himself had begun to investigate the distinctions within each stage late in his life. According to some critics (e.g., Sullivan, 1967), this sort of study was ill-advised. They claim that, like other philosophical credos. Piaget's doc- trines are not subject to strict empirical confirmation. Critics maintain that Piaget should have remained consistent. Since he began his studies in the effort to describe an elusive phenomenon in a theoretical way, he should not have tried to predict the practical applications of his theories. Nev- ertheless, as his fame grew. Piaget did begin to advise teachers on the practical uses of his ideas. There is substance to the charge made by critics. Piaget did not ex- pound the details of his theory clearly enough so that they could be investigated empirically. Careful observers have found, for example, that the degree of stage attainment is diffkuh to measure. It is possible for a child to exhibit some of the characteristics of concrete operations (e.g., conservation of number) but not others (e.g., conservation of weight). Even in the 2314 Tikeiegi Gifted Depolopterst 231 stage of formal operations, some observers have noted that it is possible to have achieved formal operations in mathematics, but not in literature (Car- ter & Kontos, 1982; Sullivan, 1967). Carter and Ormrod (1982, p. 19) conclude that "available evidence. .implies that an individual might be a formal thinker in one area but not in another." How does one judge stage attainment in such a context? Unfortunate- ly, many researchers have chosen to ignore the issue altogether. It is typical for researchers to evaluate stage attainment on the basis of the performance of just one task; and the task used to characterize a particular stage may be difficult (e.g., Brekke, Johnson, Williams, & Morrison, 1976), or it may be easy (e.g., Dimitrovsky & Almy, 1975). The question of progress through a single stage has hardly been investigated because of the lack of definition within stages. A consensus is emerging, however, about the relative difficulty of various conservation skills at the concrete operations stage, but this consensus cannot yet be interpreted to indicate conclusively the existence of an invariable developmental sequence within the stage of concrete operations. The issue is even more obscure in the stage of formal operations. Research also needs to determine whether or not progress in cognitive development is measurably continuous. It is possible that cognitive development occurs so covertly that qualitatively different stages emerge quite suddenly. A good deal of research remains to be done before the relationship of IQ and Piagetian stage development can be clarified. For the time being, Carter and Ormrod's observation that "progression within the concrete and formal operational stages takes place more quickly for gifted children than for normal children" seems reasonable (Carter & Ormrod, 1982, p. 111). This tentative conclusion, however, is little different from the observation that gifted children learn faster than average children. Perhaps the questit n, though interesting, does not hold great practkal significance for gifted education. Its relevance may be more critical to theory (Caner & Kontos, 1982).

Implications for ,ancept development.It isdifficult to draw inferences about concept development directly from the empirical research based on Piaget's theories. Even to follow the suggestions of proponents of Piaget may be unwise (cf. Sullivan, 1967). For example, many observers have noted that mathematical logic in- fuses Piaget's sense of cognitive development. This observation suggests that, if attainment of formal operations can be measured reliably, it might be possible to decide if students should begin the study of algebra at an early age. Students who have not begun to acquire formal operations in the area of mathematics might be counseled to study algebra later. Carter and Ormrod (1982) have devised a test for formal operations in the social

2 4o 232 Typical Gifted Development sciences. The relevance of this test (or similar tests) to success in algebra has not, however, been investigated.

Piaget and learning style. There is one trend in gifted education to which Piagetian notions may nonetheless be relevant in a practical way. Roeper (1966, 1978) noted that children in different stages may be expected to think differently from children in other stages. More specifically, it might be possible to think of stage attainment as "learning style." If a teacher of the gifted needs to account for learning style (a requirement in some states), then perhaps the Piagetian scheme is the most instructionally relevant way to proceed. For example, if a child does not conserve number (i.e., if the child has not attained the most rudimentary sign of concrete operations), there is some justification for proceeding to teach basic arithmetic facts using ma- nipubtives rather than drilling with flashcards. The benefits of this kind of decision to children diagnosed as preoperational are, however, largely unknown. Despite these reservations, Piaget's work provides better justification for such an action than some other theories of learning style, and it is more academically germane than some alternatives.

Cautions. One caution involves the relevance of Piaget's contributions to the study of particular educational practices. Such relevance has not been confirmed empirically. Glass (1983) suggests that an educational practice may be effective simply because a teacher believes it to be effective. (Glass does not, of course, mean that worthless educational practices can be effective.) Therefore, teachers who see a clear link between their methods of teaching, Piagetian notions, and their students' performance, may find it helpful to adopt strategies like those suggested above. Another caution involves reliable and valid assessment of stage attainment. Most tests that purport to measure stage attainment include only a small number of items (one to five items), and the reliability of such very short tests cannot be determined. Very short tests are by nature more unreliable than longer tests. For longer tests, good reliability is possible to obtain. Carter and Ormrod (1982), for example, report reliabilities in the neighborhood of 0.80 for their 30-item test. Only a few tests of Piagetian stage attainment, however, make the attempt to establish reliability. Because researchers believe it is possible, for example, to attain formal operations in one domain, but not in another, validity is an even more important unresolved issue than reliability. Reliable tests that are unrepre- sentative of the domain they intend to sample may yield deceptive results. With a good deal of work a very competent teacher could probably, over the course of several years, develop a reliable and representative test relevant to attainment of formal operations in a particular academic domain (arithmetic, algebra, history, literature). Such an effort would make a worthwhile research project and might be useful instructionally. 24i Typkal Gifted Devdopeesed 233

SOCIAL AND EMOTIONAL DEVELOPMENT OF GIFTED CHILDREN Most lists of the social and emotional characteristics of gifted children could serve as compendia of middle-rlass virtues. Gifted children are typically characterized as friendly and outgoing, independent, self-confident, honest, trustworthy, and possessed of a good sense of humor. In startling contrast to these ascribed traits are the grave concerns many educators and parents express regarding gifted children's social and emotional well-being. Among the problems predicted for the gifted are (1) difficulties in social interaction due to the differences between gifted children and their chronological age peers (Zaffran & Colangelo, 1979), (2) anxiety because of pressure to excel academically, (D'Heurle, Mellinger, & Haggard, 1959), (3) a tendency to drop out of school (Clark, 1983), (4) delinquency (Seeley, 1984), and (5) suicide (De lisle, 1986). This section of the chapter presents a brief summary of research on the social and emotional adjustment of gifted children. It also considers possible explanations for educators' and parents' anxiety about gifted children's su.:ial and emotional adjustment in the face of the group's apparent normality.

Gifted Children as "Supernormal" As we discussed earlier, Terman (1925) and his co-workers studied the social and emotional development of gifted children. One reason for this research was to disaaer how gifted children differ emotionally and socially from other children. Terman (1925) and his co-workers assessed the play 'Ind activity interests of gifted children and recorded their performance on tests of emotional stability, honesty, and trustworthiness. Ter- man's gifted group was described as normal (or better) on virtually every index' of social and emotional development. Teachers and parents rated them as more sympathetic, generous, conscientious, and truthful than other children; and their performance on tests measuring emotional stability was above the norm. Far from displaying the physical, social, and emotional problems that might be predicted from Lombroso's (1891) observations and from popular stereotypes, most of the children were healthy, happy, and popular. Most were also high achievers. Lewis Terman's longitudinal study is by far the most comprehensive study of the social and emotional development of gifted children, and other studies have tended to validate his findings (e.g., Boehm, 1962; Gal- lagher, 1958; Miller, 1956; Lehman and Erdwins, 1981). Nearly all studies have found that gifted children are, as a group, at kast as healthy, both socially and emotionally, as other children.

Social interactions.Although the difference between the gifted and a group of nonselected children was small, Terman's subjects played alone

242 sts4 ryfriecti Gifted Devektmand somewhat more. They enjoyed activities that require comparatively little social interaction, such as reading, cards, puzzles, checkers, and chess. Be- cause of these interests, about one-third to one-half of the sociability ratings of the gifted children were below the lower quartile of the control group. Nonetheless, the gifted weremoreinterested in social activities than the norm, and 84 percent of the giftedmeadthe mean of unselected children in socialinterest.The children with whom the gifted preferred to interact socially were their older playmates (Terman, 1925). Although gifted children are popularly believed to be socially inept, the basis of such beliefs may be a matter of gifted children's different values rather than their ignorance of social conventions. Based on a review of research on social cognition, Shantz (1975) concluded that advanced cognitive development is related to advanced social cognition. Scott and Bryant (1978) found that early readers in kindergarten showed greater social knowledge than did nonreaders. They also found a significant correlation between social knowledge and social behavior for these young readers. The early readers tended to have positive interactions with their peers, even though they interacted more with adults and less with their peers than did the nonreaders. Roedell (1978) also found a positive correlation between intelligence and social knowledge, butnotbetween intelligence and social behavior. Apparently very young children, like adults and older children, may be aware of social conventions yet not act in accord with them. There is no empirical evidence to support the belief that gifted children are socially incompetent. Rather, it seems that they are socially competent but that some gifted children, because of their preference for activities that engage their intellect, avoid social interaction with age-mates. They may prefer the company of older children or adults; or they may prefer solitary activities, such as reading. These preferences do not indicate emotional problems, nor do they appear to have a detrimental effect on gifted children's emotional well-being.

Emotional stability.Contrary to the predictions of many educators, few empirical studies have found gifted children to be less emotionally healthy than other children. Most studies of gifted children's levels of anxiety, self-concept, and self-confidence indicate that they are, as a group, emotionally healthy (Galluci, 1988). Feldhusen and Klausmeier (1962) found that the higher a child scored on the W1SC, the lower the child's anxiety level. A more recent study, involving a sample of over 500 gifted children ranging in age from six to 19 years, demonstrated lower levels of anxiety among gifted children than other children their age (Scholwinski & Reynolds, 1985). These gifted students were among 5,000 children who completed the Revised Children's Manifest Anxiety Scale (Reynolds & Richmond, 1985), which yields three

243 1)picat Gilled Dewlaps** 255 factors of anxiety; "physiological," "worry/oversensitivity," and "concentration." The high-IQ group scored significantly lower on all three scales. According to most research, gifted children's concepts of themselves seem to be at least as positive as those of other children. In a study of the effects of aptitude and achievement on the self-perceptions of elementary schoolchildren, Davis and Connell (1985) found the gifted group to score significantly higher on self-evaluations of competence, feelings of mastery, and preference for independent decision making. Colangelo and Pfleger (1979) found that high-school-age students identified as gifted have a higher than average academic self-concept. Tidwell's (1980) study of 1,593 high-school students with a mean IQ score of 137 found these students to have high self-esteem, a positive self-concept, and an internal locus of control. A study comparing bright students' and slow students' willingness to trust their own judgment on an estimation task found that the bright students were less influenced by other children's estimates (Lucito, 1964). As Davis and Connell (1985) point out, gifted children ought to perceive themselves as smart and capable: Most of them have a history of success in academic and extracurricular endeavors. The relatively few studies that have found gifted students to be more anxious (D'Heurle et al., 1959), to have poorer self-concept, (Freeman, 1979), or to be less well adjusted (Cornell, 1984) than other children, have been based on atypical samples. Most of these studies evaluated the self-concepts of bright underachievers or gifted students whose parents sought help because of their children's' behavioral or emotional problems. Considering their difficulties, it would be surprising if such students were found to be less troubled than average students. From Terman's earliest studies through studies conducted more recently, gifted children have met or exceeded the no.-ir on measures of emotional stability. These findings do not imply that gift& children are immune to emotional problems, but they do indicate that the gifted are no more at risk for serious emotional problems than are other children. Gifted Children as Deviant Because children are identified as gifted by virtue of their precocious intellectual development, they are by definition a deviant group. We have seen that this deviance seems to cause no serious damage to their social and emotional well-being; however, this does not mean that it has no effect at all. Giftedness brings with it disadvantages as well as advantages that can be understood in terms of harmonies and conflicts between the values of society at large and those of gifted children and their families. Most identified gifted children are members of middle- and upper-class families that subscribe to the dominant beliefs of the larger society. However, the families of gifted students value intellectual activities more than most. There

214 236 Typical Gifted Development are more books in their homes, and the parents are more involved in the children's schooling (Terfertiller, 1986). Most high-achieving children come from families that are more cohesive and supportive than other families (Cornell, 1984). These characteristics contribute to gifted children's superior achievement and socioemotional health, However, they also contribute to conflicts between gifted students and their teachers or classmates.

Effects of deviance.Many adults fail to recognize how deviant gifted children's abilities are. For example. a teacher may assert that the first grader who reads books written for fifth-grade children is reading with little comprehension (indulging in a profuse display of "word calling"). According to Cornell (1984), some adults idealize gifted children and credit them with every human virtue, whereas others resent them. Gifted children are sometimes accused of intellectual pretensions, and their parents are accused of pushing them too hard. The fact that some gifted children feel ambivalent about their deviance is not necessarily symptomatic of emotional difficulties. Their feelings may reflect society's ambivalence toward their talents. The active support of their families usually helps them to maintain their social and emotional stability in the face of others' fear and resentment. Socialized to use their superiority to best advantage, gifted students nevertheless get into difficulty if they fail to behave diplomatically ( Janos, Fung & Robinson, 1985; cf. Hollingworth, 1942, on "suffering iools gladly"). Gifted children are treated with hostility if they themselves acknowledge the difference between their abilities and those of other children. Generally, gifted children seem to heed the messages of others. In fact, Terman (1925) considered gifted children's tendency tounderstate their abilities as a sign of superior character. The modesty demanded of gifted children is, in part, due to society's discomfort with the apparently unearned advantages of giftedne 3. The term "gifted" reflects the belief that superior ability is gratuitous (Cornell, 1984). Some teachers and classmates look for weaknesses in gifted children in an effort to satisfy a sense ofjustice. They may be offended by the notion that the gifted excel by dint of good fortune rather than hard work. Thus, at the same time that they admire gifted children's abilities, teachers and classmates may resent the ease with which gifted children master material that is difficult for others. When enrichment programs are simply entertaining rather than academically challenging, as they sometimes are (Renzulli, 1977), they fuel this resentment. Such programs offer gifted students privileges rather than appropriate academic work (cf. (ox et aL, 1985).

245 Typical GsdwIojmtit 237 Emotional Problems, Real and Imagined There are several reasons for the unusual concern over gifted children's emotional welfare. One is that adults who identify closely with gifted children attribute to the children some of their own fears and anxieties (Cornell, 1984). Another is the belief that though these children are not necessarily in greater jeopardy than others, society's kw is greater if they are emotionally or socially aberrant (Gallagher,1985). Another reason for unusual concern is the belief that gifted individuals match the stereotypeof the oversensitive, unbalanced genius. Though contradicted byTerman's research, the stereotype persists because of anecdotes and case studies of famous persons who have experienced severe emotional problems. These perceptions result in unwarranted support for counseling and affective education programs for a group that is comparativelyhealthy. Nonetheless, there are few serious problems peculiar to gifted students and few gifted students with severe social or emotionalproblems. Teachers need to distinguish between real and imagined emotional and socialprob- lems of gifted students.

Real Problems of Gifted Children The problems of gifted children, as we pointed outearlier, often result from differences in values. In a society whose priorities inhibit gifted children's intellectual development, such problems are inevitable.Unfonu- nately, these problems are ascribed too often to the child's "oversensitivity" or to the child's "poor interpersonalrelationships." Some of these problems are created by parents who, whiledemanding superior achievement, do not offer the support needed to helptheir children meet high standards. Other problems are created byteachers who assign busywork to gifted children. Conflicts among peers may occur when gifted students realize that many of their classmates value sportsand fashion more highly than intellectual activities (cf. Coleman, 1961:Tannen- baum. 1962).

Peer relationships.Many gifted children feel different from other children, and some find it hard to establish friendships (Janos,Marwood, & Robinson, 1985). This is particularly true, as it is for most children,during adolescence. Society's ambivalence toward intellectual ability is reflectedin the ambivalent statements gifted children make abouttheir relationships with others. These statements can be angry and self-congratulatory atthe same time: It makes you feel like a real outsider when people turnagainst you just because they find out you're gifted. It makes you feel like a freak.

24 6 08 ripindGoaDepdopend

Just when you think you may be finding someone who understands what you're all about, you get a teacher who seems to be your worst enemy, just because you know too much! (Erlich, 1982, p.41)

Parents and teachers of gifted students often observe that these students make themselves empopular by being argumentative or opinionated. Gifted children are probably not oblivious to the effect of their pronounce- ments, but many of them choose to express their opinions anyway.Al- though many of Tidwell's (1980) highly gifted secondary school students saw themselves as unpopular, they still described themselves ashappy. Young gifted children may focus on their similarities to other children rather than their differences. Among a group of gifted students ranging in age from five to ten years old, only a little over one-third saw themselves as different ( Janos et al., 1985). There was an inverse correlation between feelings of differentness from peers and positive self-concept scores among these gifted children. The authors assumedthat the children's giftedness caused the feelings of differentness; but they had no evidence of this relationship: There was no control group to determine the extent to which nongifted children saw themselves as different fromoth- ers. All of the gifted students scored above the m.,an on a measureof self- concept, however, regardless of whether or not they thoughtof themselves as different. Nevertheless, some gifted children may have difficulty finding friends because, though they prefer older playmates, older children do not always welcome their advances. The higher the IQ, the greater the difficulty may be. iligitly gifted ,:hildren are unlikely to find age-matefriends with similar interests (Hollingworth, 1942; Roedell et al., 1980). The problem of finding friends is an important one, but it is a problem of the type that all children face in one form or another. The problem is usually resolved, often with support from the family, and is seldom associated with serious emotional maladjustment.

Overseessitiveueu.The belief that gifted children are more sensitive than other childreneven oversensitiveprobably stems from both anecdotal reports and research on gifted adults. There is some evidence that highly intelligent, creative adults are more sensitive than other adults. MacKinnon (1962/1981), for example, found some authors to be unusually aware of their own feelings. These exceptionally creative authors (who were also highly intelligentIQ 140 and above) seemed to be more aware of or more susceptible to feelingsof despair, more excitml by abstract ideas, and more subject to bad dreams thanother authors in the study. Freeman (1983) notes that oversensitiveness informs the work of creative adults, but that it makes life difficult for gifted children. There is,

1247 Typical Gifted Develepwat 239 however, little research on children's sensitiveness as it correlates with intelligence and none sufficient to substantiate the claim thatgifted children are oversensitive. In fact, as mentioned earlier, at least one study has found gifted children less likely to be oversensitive than others (Scholwinski & Reynolds, 1985).

Incidence of Severe Social and Emotional IN.adenvi In general, there is a lower incidence of severe social and emotional problems among the gifted than among the nongifted population. This lower incidence is not necessarily a correlate of high intelligence. It is more likely to be a correlate of the higher socioeconomic status enjoyed by the majority of individuals idenqied as gifted.

Delinquency.According to Ehrlich (1982, p. 41), "When we speak about problems related to giftedness, we must realize that the gifted are generally a law-abiding, conforming group that contributes comp tratively few offenders to the social order." When a high IQ score is the criteion for giftedness, the incidence of delinquency among gifted adolescents is very low (less than 5 percent). Those gifted children who are delinquent are usually from very poor homes. According to Seeley (1984), bright delinquents are even more likely than other delinquents to comefrom single- parent families, and they are more likely to be runaways. Bycomparison with other delinquent girls, those with IQ scores between 114 and 149 are more likely to be illegitimate, adopted, orseparated from their mothers (Seeley, 1984). According to Seeley (1984), many kinds of ability are represented in the delinquent population: academic ability, psychomotor ability, and artistic ability. However, delinquent students' abilities are usually unrecognized by the school system. Gifted delinquents are often severe underachievers who are more highly able in nonverbal than in verbal skills. Anolik (1979) found that gifted male delinquents were underachievers and likely to be classified as mildly disturbed. Nevertheless, they were slightly less neurotic and psychopathic than other male delinquents.

Mental disturbance.Most studies of mildly and moderately disturbed children have found their average IQ score to be in the low normal range. However, the range of scores extendsfrom retarded to gifted with relatively few disturbed children in the upper IQ ranges (Kauffman, 1985). In the severely and profoundly disturbed population, there appear tobe even fewer children with high IQ scores,theories linking genius and schizophrenia notwithstanding. "The highly intelligent, academically competent schizophrenic child is a rarity. . ."(Kauffman, 1977, p. 122). When 24 8 240 Typical Gifted Deodopaend compared with other adults, a smaller proportion of Terman's gifted group was hospitalized for mental disturbance (Terman & Oden, 1947).

Suicide,lc is popularly believed that gifted children are more likely to commit suicide than other children. There is, however, no empirical support for this belief. Studies of Terman's gifted sample found the incidence of suicide among them to be virtually the same as among the general population: For men the incidence was slightly lower, for women slightly higher (Terman & Oden, 1947). By the time The GYied Chem Grows up (Terman & Oc len, 1947) was written, five male and two female subjects had committed suicide. By 1970, twenty-eight of Terman's gifted adults had killed themselves (Shneidman, 1981). Shneidman's (1981) analysis of Terman's data on a group of men who had committed suicide found some characteristics typical of this group. These men were more likely than other gifted men to have been rejected by their fathers, to have been disturbed as adolescents, to have been mar- ried several times, and to have been alcoholic. These men were more likely to have been disappointed in themselves, unstable, lonely, perturbed, and impulsive. There was, in some cases, a wide disparity between these men's aspirations and their accomplishments. A comment included in the file of one of these men is illustrative: "My gifts, if there were any, seem to have been a flash in the pan" (p. 261). This disparity is not always the case, however. At least three of the 20 men who committed suicide were considered outstandingly successful. Other studies of suicide among intellectually superior males (Blachly, Disher, & Roduner, 1968; Paffenbarger & Asnes, 1966) have also found some common themes: early rejection by or death of the father, impaired health, depression, and moodiness. Characteristics such as these may, however, be typical of all individuals who commit suicide (see e.g., Kauffman, 1985).

SUMMARY

This chapter reviewed the literature on the typical development of gifted children. It contrasted popular views of the gifted with empirical findings in an effort to distinguish realistic perspectives of gifted individuals from stereotypes. In its review of research on the language development of gifted children, the chapter showed that talking early bears little predictive relationship to verbal talent. Nonetheless, there is evidence that children who have been continually engaged verbally by their parents perform better on cognitive tests. Early reading was found to be more common among the gifted than among other students, though by no means universal. In particular it

2 4 TYPical Gifted DevdoPmeld 241 seems that precocious reading ability may depend more on excellent decoding skills than on high IQ. The chapter found the issue of concept development among the gifted to be somewhat problematic due to methodological problems. The most prominent investigator of concept development, Piaget, did not use methods inherently compatible with the study of gifted children. Nonethe- less, other researchers have used Piaget's theories to study gifted children. Often, but not always, these researchers have found gifted children to perform somewhat better than average students on measures of stage attainment. Caution in applying the principles of Piaget was advised. The chapter concluded with a look at recent work on the social and emotional development of gifted children. Although gifted children may have some different problems from other children (for example, problems finding peers with similar interests), these problems are usually coun- teracted by the influence Or intervention of supportive homes. When serious problems do occur among the gifted, the causes appear to be similar to those associated with maladjustment in other children. The presence of exceptional intelligence does not seem to contribute to these severe problems. The intellectual superiority of gifted children may, in fact, relieve them of some of the anxieties faced by other children.

250 Precocious Development

I.Focusing Questions ll.Extreme Precocity A. Influences on Precocity 1.Social influences 2.The nature of various disciplines B. Problems of Description and Research 1.Study of mathematically precocious youth Children with Very High lQs A.Hollingworth's Research 1.Child A 2.Child D 3.Child E 4.Common characteristics IV.Extreme Mathematical Precocity A.Study of Mathematically Precocious Youth 1. Cognitive characteristics 2.Affective characteristics 3. Sex differences 4.Creative behavior S.Values 6.Vocational interests 7.Follow-up studies B. Other Research 1.Problem-solving strategies V.Musical Precocity A.Characteristics of Famous Musical Prodigies 1.Early identification and instruction 242 251 Precocious Development 24$

2.Precocity and eminence 3.Development of virtuosi VI.Literary Precocity A.Juvenilia of Famous Authors B. Developmental History of Precocious &it! C.Child Authors VII.Athletic Precocity A.Development of World-Class Tennis Players B. Development of Olympic Swimmers VIII.Other Precocity A.Eidetic Imagery (Photographic Memory) B. Rapid Arithmetic Calculation C.Chess Playing IX.Summary

Focusing Questions

1. Why must researchers rely primarily on case studies to examine the characteristics of extremely precocious children? 2. In what ways have the parents of prodigies, such as John Stuart Mill, Michael Grost, and Terence Tao, been influential in guiding their children's education? 3. What cognitive and affective characteristics are associated with extreme mathematical precocity? 4. How do the families of musical, athletic, and literary prodigies support the development of their children's talents? 5. What relationships appear to exist between intelligence and specialized skills such as eidetic imagery, rapid arithmetic calculation, and expert chess playing?

EXTREME PRECOCITY

Extremelyprecocious children are as different from typical gifted children as typical gifted children arefrom the average. Sometimes that difference intellectual giftedness is usually associated with IQ scores that are atleast four standard deviations above the mean. Th,:incidence of such high scores is about 3 in 100,000. Nevertheless, extreme precocity is quite rare. For example, extreme intellectual giftedness is usually associated with IQ scores that are atfour standard deviations above the mean. The incidence of such high scoresis about 3 in 100,000. Influences of Precocity In the popular imagination, child prodigies arefreaks of nature, children whose innate abilities set them apart from others.Prodigious development, however, may not relate to innate factors.Some authorities attribute 25 2 244 Prerodows Devdopsand prodigious development to social factors, others to the demands of different disciplines. The fact that more child prodigies have been identified in certain disciplines can be interpreted in either of these ways. Some writers, however, attribute prodigious development to a complex interaction of personal and social influences (see e.g., Feldman, 1979).

Social influences.The role that society plays in children's prodigious development is not easily determined, and seve.al different views have been put forth. The same social variables that affect the achievement of other individuals seem to influence prodigious development: social class, race, sex-roles, ethnicity, and birth order. The majority of prodigies are boys, usually the first-born or only sons in middle-class familieskMacLeish, 1984). Even geographic location can affect the development of precocious talents. Fifty percent of the chess prodigies in the United States come from three metropolitan areas that contain about 10 percent of the U.S. population: New York, San Francisco, and Los Angeles (Gardner, 1982). Pressey (1955) relates the different numbers of prodigies to the values society places on different disciplines. According to Pressey (1955), finan- cially rewarding disciplines, such as professional athletics, are more likely to produce prodigies than less rewarding ones, such as painting. Simonton (1984) suggests, however, that social rewards may actually inhibit the later productivity of child prodigies. Once an individual has reached a goal, his or her motivation may lessen; and acclaim may bring social obligations that are time-consuming.

The nature of various disciplines.More prodigies have been identified in music and mathematics than in other academic or artistic fields (MacLeish, 1984). Some theorists see this difference as a function of the disciptme. According to Page (1976), certain disciplines depend mote on manipulation of symbols than on access to a large amount of knowledge; these disciplines are the ones in which prodigies can infer much from a relatively small set of examples. Disciplines, like history, that depend on accumulation and analysis of a great deal of knowledge may require years of study. This reasoning suggests that the more a work depends on the use of factual information, the longer it takes to accomplish. In general, the peak period of the careers of eminent persons seems to be between the ages of 30 and 40 (Simonton, 1984). This early peak may depend on prodigious development in childhood (Gardner, 1982). Differ- ent disciplines have slightly different curves. In history and some other scholarly fields, the peak productive age is late, in the sixties. For poetry. music, and drama, the peak is in the thirties Or forties. The peak period of scientists is earlier still. According to Simonton (1984, p. 99), "Scientific geniuses of the highest historical order tend to begin their careers at unusually early ages." 25 3 Precocious Development 245 Problems of Description and Research There are very few studies of highly gifted children or prodigies because there are very few prodigies. Finding a large number of prodigies for study is difficult, and using parametric statistics in research with small sample sizes is inconclusive. Therefore, educators should be cautious in applying evidence from the studies that are available. Foster (1986) supports certain uses of single-subject research, however.This type of research is descriptive and is useful to illustrate some similarities and differences among the highly gifted. Unfortunately, much of the literature about precocious children is inaccurate and reflects common stereotypes of giftedness. Some of the literature claims that such children have serious failings that cot nter- balance their exceptional intellectual or artistic development. Usually, this literature claims that extremely precocious children are eccentric, unbalanced, or overly specialized. A comment about William Sidis exemplifies this type of reasoning. "He was brilliant at mathematics, but in all other subjects he was childishly ignorant. . ." (Huxley as quoted in Wallace, 1986, p. 118). This statement conflicts withbiographkal information about Sidis. According to Wallace (1986), for example, Sidis was so intelligent inail areas that he qualified to enter Harvard atnine years of age.

Study of mathematically precocious youth.Since the early 1970s. researchers at the Johns Hopkins University have been identifying andstudy- ing brilliant young mathematicians (Stanley, 1976a). Becausethis project entails an extensive talent search, it is able to locate relatively largenumbers of such students. Consequently, the findings from the SMPYresearch are more broadly applicable thanfindings from case studies.

CHILDREN WITH VERY HIGH IQS

A number of researchers have investigated the characteristicsof children with extremely high 1Qs. Hollingworth (1942) assembled casestudies of children with 1Q's of 180 and above. Terman compared a group of very high IQ children with the other gifted children in his sample(Terman Sc Oden, 1947), and Jenkins (1943) collected biographical informationabout black children with 1Qs alx 160. Hollingworth's (1942) investigation of highly intelligent children is the most comprehensive of thesestudies.

Hollingworth's Research Hollingworth (1942) fbllowed the progress of twelvehighly gifted children from early childhood through young adulthood.The fbllowing examples of behavior characterize the children in this study.

25 4 244 Precede= Develepasst

Child A.This precocious child began to talk at 10 months of age. He could say the alphabet forward at age 12 months and could say the alphabet backwards at age 16 months. Child A began to read at three years of age and was reading fluently by the time he entered school. He was also talented in music and in rapid arithmetic calculation. When he was six years old, Child A scored a mental age of 12 years 2 months on the Stan- ford-Binet Intelligence Scale. In spite of his high ability, he did not earn exceptionally good grades, and his teachers were not very .mpressed by his talents.

Child D.Like child A, this child also learned to read long before school age. In his preschool years, Child D amused himself by constructing a language, history, and literature for an imaginary culture. Throughout early childhood, he enjoyed composing music and was talented at playing the piano. By the age of 12 years, Child D had completed college entrance requirements in algebra, geometry, and trigonometry. He was admitted to college at the age of 12 years 6 months; he graduated with Phi Beta Kappa honors at the age of 16 years.

Child E.By the time he was 8 years old, Child E was able to speak and understand simple sentences in four or five languages. At this age, his knowledge of English vocabulary was equal to that of an average adult. When he was 10, Child E began to do original historical research in the classics. He continued this activity throughout the remainder of his years in school.

Common chanwteristies.The children that Hollingworth studied had a lot in common. They read before they entered school; and with few exceptions they achieved well in school. Most attended college early and many graduwed with honors. These children had intellectual and artistic interests that persisted throughout their school careers. Other biographical and autobiographical records confirm these generalizations (see Box 9-1). Many eminent individuals, however, lived before IQ tests weir used. Biographers sometimes estimated their Kas from information about their developmental histories (see e.g., Cox. 1926). The validity of ISis practice is questionable. Nevertheless, biographies of children whose high-lQs are known (see e.g., Grost, 1970 ) also confirm Honing- worth's findings.

BOX 9-1Prodigiesand Their Education

JOHN STUART MILL The degree to which Mill's exceptional achievement resulted from innate talents or from the extraordinary education he received from his father is unknown. Nevertheless, Mill accomplished prodigious feats at extremely early ages. By the age of eight, for example, 255 Pneeciese Development 247 he had read, in the original Greek, all of Herodotus, part of Lucian, and several of Plato's dialogs. James Mill's educational methods, though apparently highly effective, were controversial during his time. They continue to be a subject of controversy today. Kelly (1 985), for example, alternately lauds James Mill for effectively developing his son's intellect and blames him for harming his son's social and emotional development. At one point, she goes so far as to suggest that, had he been living in our times, James Mill might have been jailed for child abuse. The principal characteristics of John Stuart Mill's schooling were as follows:

one-to-oneinstructionat home under his father's tuition, early echooling in the classics, extensive reading and discussion, emphasis on critical thinking in the context of academic disciplines, exceraionally high standardsfor intellectual achievement.

MICHAEL GROST Michael Grost is among the more recently acclaimed prodigies. His IQ score has been estimated to be over 200. Audrey Grost'sGenius in Residencedescribes her son's schooling. After many efforts to obtain an appropriate education for Michael in the public schools, his parents took him out of elementary school and enrolled him in college. He was ten years old. This measure was not so drastic as it seems. It was preceded by his having taken several college courses on a part-time basis and earning an *A" average ("Where Are They Now?" 1969). During his years as an undergraduate, Grost won several awards and fellowships despite being much younger than the other competitors. By the time he was 16 he was enrolled in a doctoral program at a major Ivy League university.

TERENCE TAO Terence Tao received his education in Australia. His exceptional mathematical giftedness was recognized at a very early age: He was the first eight-year-old to score 760 on the SAT-M. This score placed him in the 99th percentile for high-school juniors and seniors. Although Terence enteml school at the usual age, his father soon saw the need for him to receive special instructional provisions. At the age of 6 and a half, Terry was receiving instruction in the third, fourth, sixth, and seventh grades. When he was 8 and a half, Terry began high school full time. There, too, he was placed in different grade levels for different subjects: Grade 12 for physics, Grade 11 for chemistry, Grade 10 for geography, and Grade 8 for general studies and homeroom. At the age of 9, Terry began taking some university courses in physics and mathematics. His father plans for him to continue in a dual plac-ment until he is at least in his early teens. (For more information, see Tao, 1986.)

EXTREME MATHEMATICAL PRECOCITY

Since 1971, researchers at the Johns Hopkins University have been studying mathematically precocious youth. Their research considers the cognitive and affective characteristics of these highly talented students, and it

256 248 Precorims Deveispnest traces their longterm progress. In addition, by testing tens of thousands of able students every year, SMPY is able to locate relatively large numbers of very talented young mathematicians.

Study of Mathematically Precocious Youth The researchers at Johns Hopkins have been interested in studying mathematically precociousstudents, in contrast to students withaptitudefor mathematics who do not manifest their talents at early ages. According to Keating (1976b, p. 24), "Precocity... meansarriving at some stage of development earlier than expected, such that the individual's current state of development is more like that of someone much older." Quantitative aptitude, on the other hand, may entail sophisticated abstract reasoning without rapid learning (Stanley, 19766). Precocity in mathematics involves early aptitude and interest in mathematics, but it does notnecessarilypredict later prodigious achievement in mathematics or science (Keating, 1974). On the other hand, some creative mathematicians and scientists do not demonstrate precocious achievement early in life.

Ctwnitive characteristks. The most significant cognitive characteristic of mathematically precocious youth is their rate of learning. According to Stanley (1976b), this characteristic is relaied to verbal aptitude. Hence, most students who demonstrate mathematical precocity also have superior scores on tests of verbal intelligence. High abstract reasoning scores on nonverbal intelligence tests, such as Raven's Progressive Matrices, are not as successful in predicting learning rate. In general, results from SMPY suggest that precocity in mathematics is associated with superior academic achievement. Students who show early talent in mathematics seem to maintain their advantage over other students. Even among the group of talented young mathematicians, the trend seems to hold: The most talented maintain their advantage over the less talented. A number of studies have compared the academic performance of mathematically precocious students with average and above-average peers. In a follow-up study of talent search winners, Benbow (1983) found that these students did significantly better than other college-bound students on measures of academic achievement: Their eleventh- and twelfth-grade SAT scores were higher, they took more science and mathematics classes, their grades were better, they entered college earlier, and they attended more academically challenging colleges. A study comparing the most talented with the less talented SMPY students (Benbow, Perkins, & Stanley, 1983) showed that the most talented group outperformed the less talented on a high-school administration of the SAT. The most talented students in the SMPY group also took more

25 7 Prevociess Dsreirpsest 249 science and mathematics courses than their lesstalented counterparts; they performed better on the CEEB mathematics achievement tests;they scored higher on the National Merit Scholarship Exam (the PSAT);they participated in more mathematics and science contests; theychose accelerative options more frequently; and they were morelikely to choose a college major in a mathematics or science field.

Affective chara..teristics.Haier and Denham (1976) evaluated the nonintellectual characteristics of 71 mathematicallyprecocious students. Their research confirmed the findings of an earlierstudy (Weiss, Haier, & Keating, 1974). By administering severalstandardized tests of affective performance to the students, the researchers were able to comparethe mathematically gifted students to average peers and toadults. The tests administered were the California PsychologicalInventory, the Eysenck Personality Inventory, the Study of Values,Holland's Vocational Prefer- ence Inventory, and theAdjective Checklist. The results of the study enabled the researchers to conclude, "As a groupthese mathematicahy precocious youth are interpersonally effective andsocially mature. .Both boys and girls are confident and well-adjusted ..." (Haier & Denham, 1976, p. 239). The California Psychological Inventory (CPI)provided a measure of the interpersonal effectiveness of the mathematicallygifted students. These students were compared with age-mates, bothgifted and nongifted, and with high-school students, both giftedand nongifted. Their scores were most similar to the other groupsof gifted students and least similar to the group of nongifted age-mates. The findings suggestthat the mathematically precocious students were capable of' maturesocial interactions. This research also found that the mathematicallygifted students used their intelligence in innovative ways to improve theiradaptation to environmental conditions. In spite of their demonstrated social ability, themathematically gifted students were classified as introverted on theEysenck Personality Inven- tory (EPI). The followingadjectives characterize introversion on the EPI: "introspective, serious, planning, fond of hooks, andreserved except with friends" (Haier & Denhan, 1976, p. 234). It seems thatalthough these gifted students were able to interact with others,they preferred solitary intellectual pursuits. Results from the Study of Values supportthis finding. The mathematically gifted students dern,,ostrated a strongorientation toward theoretical values. Their high regard forknowledge and truth fit in with their preference for indepeadent,scholarly activities. Similar responses were recorded on Holland'sVocational Preference Inventory. More than half of the mathematicallygifted students favored investigative occupations over artistic, realistic,enterprising, conventional, or social occupations.According to Holland (1973). vocationalpreferences 258 250 Precocious Dew knot* reflect personality types. The mathematically gifted students most resembled the investigative personality type, which is characterized by scholarship, intellect, self-confidence, independence, caution, precision, introversion, reservation, and rationality. On the Adjective Checklist, the mathematically gifted students selected adjectives to describe themselves. Their selections tended to include more unfavorable adjectives than favorable ones, a finding that might indicate that the students had inadequate images of themselves. An analysis of the specific adjectives that were most often selected, however, suggests that the students may have been realistic in their self-appraisal. They most often selected the following positive adjectives to describe themselves: intelligent, capable, adaptable, logical, honest, and clear-thinking. They tended to choose the following negative adjectives: argumentative, sarcastic, impa- tient, opinionated, and cynical. These negative adjectives are relatively mild in comparison to some of the others that the students might have selected, and they tend to reflect the frustrations that are associated with giftedness. According to Haler and Denham (1976, pp. 238-239), the selection of these negative adjectives "may indicate the degree to which Phe students] are bored or dissatisfied with conventional school curricula."

Set dWerenees.In Chapter 6 we discussed the fact that boys get higher scores than girls on tests of mathematical aptitude. Among highly gifted young mathematicians, the trend is even more obvious (Fox, 1976b), When they used a criterion score of 600 on the SAT-M to identify mathematically precocious students in junior high schools, researchers found that many more boys than girls qualified. In a sample of 396 students who took the SAT-M, 22 percent of the boys and only 2 percent of the girls obtained scores of at least 600. According to Astin (1974), none of the girls scored above 600. Nineteen percent of the boys, however, scored above 600, and one scored as high as 790. Weiner and Robinson (1986)obtained similar results. They reported the ratio of boys to girls scoring at or above 600 on the SAT-M as 2.8 to 1. In the 1980 to 1983 study of junior high-school students with scores between 700 and 800, 269 (92 percent) boys were identified as compared with 23 (8 percent) girls (Stanley, 1985). The discrepancies between the performance of boys and girls also become more pronounced as the students get older. In a test-retest study of scores on the SAT-M in seventh and eighth grade administrations, bays gained an average of 66 points and girls gained an average of 56 points (Asfin, 1974). Similar differences occurred in the mathematics achievement test scores of boys and girls. Weiner and Robinson (1986) reported that the mathematics achievement of the students in their study was predicted better by the SAT-M for boys and by the SAT-V for girls. This finding suggests that talented boys and girls may use different cognitive strategies to learn math.

25 Precocious Development 251

These researchers found similarities in theaffective characteristics of mathematically talented boys and girls. Astin's(1974) study, however, re- vealed some differences in their careerinterests, their liking for school, the ages at which they firstbecame interested in mathematics, andtheir position in the family. She also noted similaritiesin the students' socioeconomic background, their mothers' level of achievementmotivation, and their personality characteristics (Astin, 1974). Fox (1976b) discussed the significantdifferences between boys' and girls' early involvement with mathematics.Even among precocious young mathematicians, boys reported more often thangirls that they engaged in mathematics activities outside of school.According to Fox (1976b, p. 185), "very few girls study mathematicsindependently in a systematic way, nor do many girls frequently read mathematicalbooks or play with mathematical puzzles and games in their leisure time."

Creative behavior.Studies of the creative potential andbehavior of' mathematically precocious students areinconclusive (see e.g., Keating, 1976a; Michael, 1983). According to Michael(1983), the results of these studies are confounded by theoretical andmethodological difficulties. He noted the following problems with the SMPYresearch on the creativity of precocious young mathematicians: (1) inability toconceptualize relevant subconstructs of creativity, (2) failure ofcreativity tests to predict later accomplishment, (3) use of' follow-up questionsthat did not elicit relevant information, (4) unreliability of measures ofcreativity, (5) comparison of subjects in a restricted range, and (6)violation of statistical assumptions in data analyses.

Values.Using the Allport-Vernon-Lindzey Studyof Values (SV), Fox and Denham (1974) assessed the valueorientation of 35 of the highest scoring males in the talent search. They measuredvalues on the six scales of the SV: (1) theoreticalvaluing intellectuA,scientific, and philosophical pursuits; (2) economicvaluing practical andmoney-making pursuits; (3) aestheticvaluing grace, symmetry, and artisticpursuits; (4) socialvalu- ing people; (5) politicalvaluing power;and (6) religiousvaluing mystical experiences. The researchers found that the studentsscored highest on the theoretical and aesthetic scales. Their scores onthe economic, social, and political scales were similar to average scoresfor high-school males. Their scores on the religiousscale were rated as "outstandingly low"(Fox & Denham, 1974, p. 163). Later research with a larger and lesshighly talented sample (Fox, I976c) reported somewhat differentresults. Boys in this group scored highest on the theoretical and political scales,whereas girls scored highest on the social andpolitical scales. According to Fox (1976c),girls' mean Precovisen Devekpasas scores on the social, aesthetic, and religious scales weresignificantly higher than the means for the boys. The boys' mean scores on the theoretical, political, and economic scales were significantly higher than the means for the girls. Conclusions from this research are tentative. It appears, however, that differences in value orientations may relate to gende and degree of talent (Fox, 19760.

Vocational interests.Fox and Denham (1974) studied the occupational interests of seventh- through ninth grade participants in the talent search. Using the Vocational Preference Inventory, they found that more than half of the boys and more than a third of the girls preferred investigative occupations such as engineer, doctor, and scientist. Girls most often listed "mathematics teacher" as their first choice, whereas boys most often listed "scientist." When the data were analyzed by age, the researchers found that seventh-gradz isiris rated artistic vocations most highly. They also rated social occupations more highly than the boys did. This trend was evident in the data for the eighth- and ninth-grade girls as well. Even though more than 20 percent of the girls chose social occupations, almost none of them selected "homemaker." Instead, they favored careers in teaching and other social serviceprofessions. Among the most highly talented girls, however, investigative careers were ranked highest much more often. A later study using the Strong-Campbell Interest Inventory (Fox, Pasternak, & Peiser, 1976) yielded similar results. 'the scores of both boys and girls were higher than those of average ability students on the scales that measure interest in 4westigative careers. Many of the girls in this study also tended to choose artistic and social careers. According to the researchers, "girls scored significantly higher than boys on the following interest scales: domestic arts, art, social service, music/dramatics, teaching, writing, nature, office practice, religious activities, and medical service" (Fox et al., 1976, pp. 247-248). Boys, on the other hand, scored higher than girls on the scales measuring interest in mechanical activities and science (Fox et aL, 1976). Follow-up studies. The longitudinal studies that have been conducted thus far report the college achkvements of the mathematically gifted students. According to Benbow (1983), over 90 percent of the SMPY students were attending collew at the time of the first follow-up study. In general they were attending selective four-year colleges. Their majors were most often in the areas of science,mathematics, and engineering. Sixty-one percent of the males and 50 percent of thefemales were planning to major in these fields. 261 Precocious Devdoinuent 253 These students also had high expectations for their future performance. More than 96 percent planned to get atleast a bachelor's degree; 37 percent planned to complete a doctorate.These figures should be compared to those for average high-school students, only 51 percent of whom aspire to a bachelor's degree (Benbow, 1983). Stanley and Benbow (1983) reported the college attainment of some of the students who had the highest SAT-M scores. These students entered college very early and were successful there. One of the students completed a bachelor's of science degree inmath,matks at Boise State University one month after he turned 12. Another was pursuing, at 13 years of age, a double major in mathematics and computer science at UCLA.

Other Research Some agencies have initiated programs similar to those offered through SMPY. The state of Illinois, for example, began a pilot program in 1977 and expanded its identification and special classes in subsequent years (Van Tassel-Baska, 1983). Because these programs are more recent, however, they have not yet accumulated andanalyzed sufficient data for comparison with SMPY findings. Other research, more similar to that conducted by Krutetskii (see Chapter 6), has attempted to describe the problem-solving strategies used by highly talented young mathematicians.

Problem-salving strategies.According to Wagner and Zimmerman (1986, p. 246), highly talented math students use moresophisticated strategies to solve problems than do their less talented peers.Gifted math students seem to be both flexible and competent whenattempting complex math tasks. The researchers found thatmathematically talented students were adept at the followingproblem-solving tasks:

1. organizing material. 2.recognizing patterns or rules, 3.changing the representation of the problem and recognizing patterns and rules in the new representation, 4.comprehending very complex structures and working within these structures, 5.reversing processes, and 6.finding (constructing) related problems.

MUSICAL PRECOCITY

The most famous musical prodigies arethose who have continued to produce extraordinary music as adults: singers, such as JennyLind; jazz musicians, such as Louis Armstrong;and violiniits, such as Yehudi Menuhin. Most child prodigies do not become eminent composers or performers,however. They are, nonetheless, interestingbecause of their early manifestation of ability.

26 2 254 Precocious Demist One*

Characteristics of Famots Musical Prodigies Mozart is the most famous of the many composers who were recognized as child prodigies. 1-16 abilities exemplify those of other children who became eminent composers. Beethoven, for example began to play original compositions at the age of 12; Mendelssohn, by the time he was 12 years old, had composed a piano sonata, a violin sonata, fugues, motets, and operettas; Paganini wrote his first sonata when he was eight. Mozart's general musical virtuosity, at seven years old, would have been remarkable in a grown man (Brockway & Weinstock, 1958). His manner of composing suggests that he had an exceptional memoryfor music: He apparently worked out his compositions mentally before writing them. It is reported that he could write down, note for note, a long piece of music on first hearing (Schonberg, 1970). He also hadabsolute pitch, which he demonstrated at appearances arranged by his father. Like Mozart, child composers seem to subscribe to a strong work ethic and possess an unusual love of music. Unlike Mozart, however, most contemporary musical prodigies do not compose. Technical proficiency in music tends to precede creative expression (Bloom & Sosniak, 1981). In the eighteenth and nineteenth centuries, however, the public expected that virtuoso performers would also compose. Today, many precociousclansical musicians never develop the interest or the capacity to create excellent musical compositions. The modern public expects to hear excellent performances of classical works, and training emphasizesexecution rather than composition.

Early identifscation and instruction.Parents of prodigies often recognize their childrens' talents and provide special instruction for them. Artur Rubinstein began studying with a masterteacher by the age of three; jascha Heifetz by the age of five (Sward, l 933a).The instruction of prodigies is usually expert and reflects students' rapid rate oflearning. It presses them toward increasingly difficult accomplishments(Bloom, 1982; Pressey, 1955). The teachers of musical prodigies are,almost without exception, master teachers whoprovide models of exceptional skill anddedication (Bloom, 1982). Practice in the field must begin early and become increasingly intensive in order for -utstanding talent to emerge (Bloom, 1982; Bloom & Sosniak, 1981; Sward, 1933a). Studies of prodigies indicatethat instruction, practice, and performance take up increasingly more of the prodigies' time, until eventually most of their waking hours are spent in activities related to practice Or performance (Bloom, 1982). Students show a marked commitment to the field of their interest (cf. Hollander, 1978). Willingness to work and enjoyment of work areevident (Feldman, 1979). 2 3 Prtcociatu Develop:new 255 Another important condition of achievement in music is thepremium placed on accomplishment by the children's families. Often thefamilies value music so much that they are willing to make large expendituresof time, energy, and money to further the students' education (Bloom,1982). Usually such families enjoy and participate actively in the field inwhich the children excel. Pressey (1955) remarks on the increasing recognitionand social stimulation given prodigies by family, friends, ane others whoshare the children's enthusiasm for music. Attention and praise frompeople who are significant in a person's life arepowerful reinforcers for any human behavior; it seems unlikely that attention and praise are withouteffect on a prodigy's behavior. Whether because of social rewards, innate qualities, orinternalization of family values, prodigies show an amazing willingness towork for long hours, and they take much pleasure in their work. By thetime they are adolescents, they spend more time being instructed, practicing,performing, and attending performances than in any other aspectsof their lives (Bloom, 1982). At least ten years of devotion to the studyof music preceded eminent status in all of the musicallytalented adults studied by Bloom. According to Tannenbaum (1983), this willingness todirect energy and effort toward accomplishment in an area is a highlysignificant variable in determining outstanding achievement.

Precocity and eminence.According to Sward (1933a), many people who attain eminence in music were recognized astalented before their teenage years: Of over 400 adultsidentified as exceptionally talented in musk, 83 percent of the males and 92 percent ofthe females were identified before they were ten years old. Almost half the group wererecognized as talented before they werefive years old. Sosniak (1985), however, notes that in the cases she studied, children'sexceptional musical talent was not at first evident. Only as the children's talent wasnurtured and as it matured, did their parents recognize it as significant. It is not dear, however, why some childprodigies become excellent adult musicians and others do not. No doubtthere are both intra-individual and educational factors involved. In many cases,prodigies do not receive the musical education they need torealize their potential. An extreme instance of this is reported in the historical caseof a slave called "Blind Tom," who, when he was five years old,played from memory the musical pieces his slaveowner's daughters werelearning. The child could reproduce immediately upon a single hearing,pieces played on the piano. Without any formal training, he also composedmusic. His owner began exhibiting Tom when he was eight years old.Despite his talent, Tom was denied an education that might have enabledhim to integrate this ability with other skills and to become a mature composer(Fisher, 1973).

.214 256 Precocious Development Sosniak (1985) believes that children who become successfulprofes- sional pianists receive a great deal of assistance fromtheir families. In her view, musical prodigies and their parents developtogetherlearning about the musical world together and reinforcing eachothers' expectations and actions. Bamberger (1986) has studied the cognitive development of musically gifted children to try to discover individual differencesthat might explain why some go on to become noted adult performers andothers do not. She investigated the possibility that changes in cognition, ratherthan changes in affect or lack of appropriate education, play a causalrole. She contrasts young children's performance withthat of adolescents. Young children demonstrate:

.. an ability to move their attentionfreely among the complexity of inter- secting musical dimensions Isuch as pitch, durations, melodic andharmonic functions] that together give unique coherence to even a single momentin a composition ...But in adolescence, multiple dimensions and their internal representations... come apart. ...In practicing, focus shifts among these now differentiated dimensions. .. As aresult, the functional reciprocity within the network of representations that so characterizes younger per- formersand the easy, all-at-once imitation it made possibleno longer works. (Bamberger, 1986, pp. 388-389)

Bamberger (1986) speculates that some children are unable eve,- to return to a more integrated approach to performance.More researcl necessary to confirm her speculations. Even if children's abilities keep pace with therequirements of musical performance. talented children may nonetheless fail to achieve atthe highest levels. Achievement of eminenceapparently requires years of continuous effort. Competing interests mayencroach on practice time and may diminish the child's interest in music as a career. Somechildren may find that they are not temperamentally suited to performing eventhough they love music. In other cases, lack of financial resources mayprohibit entry to the competitions and summer camps that contribute tothe pursuit of music as a career. Manyelements must combine in order for eminence tofollow precocity (Albert & Runco, 1986).

Development of virtuosi.Sosniak (1985) conducted a study of 21 prodigious concert pianists under 40 years of age. Eachof these musicians had been a finalist in a major international pianocompetition. Sosniak's study identified similar characteristics in the subjects'homes and training. The importance of the . -rents' role is evident.All of the pianists' parents arranged for them to startmusic lessons early: as young as three, and no later than nine years old. During these first years,the social rewards for their musical performance increased. As thechildren improved, they 265 Pricotions Develop mist 257 were given attention not onlyfrom their parents, but from their music teachers, and as they began to give recitals and schoolperformances, from their classmates and others. About half of the pianists' parents had at least some formaltraining in music, and most of them valued music and listened toit frequently in the home before their children began lessons. Four sets of parentsearned their living as professional symphony orchestra members andby giving music lessons. Five sets of parents were amateur musicians. Thesubjects reported that music was often played in their homes; somementioned going to sleep or waking up to music.However, the percentage of musically oriented families among these highly gifted musicians is notnecessarily much higher than average. According toSosniak (1985), a market survey in 1974 found that almost 44 percent of the householdspolled included an amateur musician; and about 79 percentreported that they listened to records often. Possibly, musicians' families were moreinvolved with the classical tradition. In most cases, the parents insisted on lessons as partof the child's education; and they insisted on regular practice.They also paid attention to the suitability of theteacher for their child and were quick to change teachers if they felt the teacher did not have apositive influence on the child's musical development. When the children first started taking lessons, they wererequired to practke about 45 to 90 minutes a day, on the average,six days a week. Practice time incrrased to about two to four hours perday. This increased involvement was sometimes required by a new teacher;but sometimes it reflected a student's growing commitment to music, andcaused the parents to seek a new teacher. Later,when the children began to study under master teachers, practicetime ranged from four to seven hours a day. Practice became a part of these musicians' routine veryearly in their lives; other activities were organized around their practicesessions. As this practice time increased, the young music students wereexcused from some household chores because their practice was consideredcrucial. They were considered "special" in a significant way and, attimes, enjoyed privileges and attention that balanced the rigor oftheir daily routine. Study with a demanding master teacher was anessential element of the training of each of these pianists. By the timethe student began lessons with the master teacher, who was, in all cases, amongthe best practicing or retired concert pianists, the student was thoroughlycommitted to music as a career. The masterteachers exerted an enormous influence on the students, who were in awe of them. They pushedthemselves harder than ever to meet these teachers'high standards. After several years of lessons, the master teacher's role became moreand more that of a coach, giving advice and making suggestions rather than givingdirect instruction. The importance of societal influences on theseaspiring concert pi- 266 258 ij Development anists was not studied by Sosniak (1985); however, she notes that there were relatively few female pianists and noblacks in the group. Thirty-five percent of the subjects came from professionalfamilies, 45 percent from white-collar families, and 20 percent from blue-collar families.

LITERARY PRECOCITY

Although few writers have been recognized as child prodigies, many wrote brilliantly in their early years. This finding seems to be particularly true of poets; there are, however, some novelists whopublished in their teens. Nevertheless, many children who write well and are prolific do not continue to produce in later life.

Juvenilia of Famous Authors Burks, Jensen, and Terman (I 930) studied the juvenile works of ten famous authors. They developed a scale for comparing these works to the literature produced by adult authors. According to their ratings, poems written by Byron at the age of 15 were slightly beuer than the average work appearing in literary magazines such as Poetry and Adantic Monthly. A poem written by Coleridge when he was 15 was considered as good as the average work in popular magazines such as Saturday Evening Post and Good House- keeping. The methodology of this study is quite naive, and the results should be treated cautiously. The researchers were not literarycritics, and the superiority of selected juvenilia does not ensure the superiority of all the juvenilia of a given author. In general, the results of this study do suggest that the childhood work of eminent authors is alot better than that of average children and adults. Some authors produced extraordinary works in their teens. Mostof Arthur Rimbaud's work was written before he was out of his teens. Percy Bysshe Shelley's "Queen Mab" was published when he was 18, and Mary Shelley started to write Frankenstein when she was 18. Other writers who produced significant works at very young ages include Blake, Boccaccio, Byron, Chaucer, Milton, Pope, and Robert Louis Stevenson.

Developmental History of Precocious Authors The study by Burks, Jensen, and Terman (1930) also reported what v...as known about the development of the tenfamous authors whose juvenilia they evaluated. It described their family backgrounds, theirearly interests, and their youthful accomplishments. According to this research, most of the writers came from middle- or upper-class backgrounds. Their parents tended to be well educated and

2 6 7 Pro:scam Arosiopimmt 259 supportive of their children's literary efforts. Robert Browning's father, for example, was reported to have beena collector of books and a strong supporter of his son's literary interests. Most of the authors displayed early interest in reading and writing. They were well-read at earlyages, often in more than one language. They also engaged in other creative endeavors suchas drawing and playing musical instruments. Keats, for example, demonstratedan early interest in poetry and nature, and he translated the Aeneid into prose during his fifteenth and sixteenthyears. Some of the authors achieved recognition when they were quite young. Keats received many prizes and medals during his school years for academic work and for literary productions. Sometimes the writers' early accomplishments defied social convention. Shelley, for example,was ex- pelled from Eton because he wrote and published an essay on the necessity of atheism.

Child Authors Another focus of the 1930 study of Burks and his associates was an assessment of the development and work of talented young writers in Terman's (1925) sample of high-1Q students. The investigators rated many of these students' poems as highly as those written by the famous authors at comparable ages. It seems unfortunate that even the best writers in the study typically did not develop their talents in writing. Except for one, they tended to enter other fields and to write only for their own pleasure. One subject, however, did become a published writer of fiction (Terman & Oden, 1947). Terman and Oden (1947) were distressed by the fact that the girls who demonstrated exceptional literary talent failed to produce significant works in adulthood. Another well-known example of this phenomenon is recorded in stories about Daisy Ashford, a child who wrote a best-selling novel at the age of nine. Her novel, The Visiiers, sold hundreds of thousands of copies over several decades, yet its young author stopped writing during her teens (MacLeish, 1984). Although the young writers studied by Burk and his associates (1930) did not become famous in most instances, their backgrounds were very similar to those of the famous writers. Their parents were well educated and affluent, but perhaps they did not have so strong an interest in literary work as the parents of the famous authors. The children showed an early interest in reading, writing, and the arts. They won competitions for their writing and had work published in children's magazines. For example, one of the girls, by the age of 16, had written many lyric poems, a novel, and several short stories and plays.

26.8 260 Precocious Development

ATHLETIC PRECOCITY

Since athletic precocity depends on physical, rather thanintellectual, development, it seems unlikely that childrencould surpass adults. However, by the time they are 12 years old, a few children can outperform evenprofes- sional athletes in some sports, such as gymnastics andtennis. In such sports, where coordination, flexibility, andbalance are most important, the differences in size and strength between children and adults do notdecide the outcome. Extreme precocity in athletics seems to require the samedevotion to practice as it does in scholarly or artistic disciplines. Childrenwho excel at very young ages have already spent yearsdeveloping their skill. World- class tennis players whose extreme talent was not demonstratedin childhood began to take lessons in the sport sometimebetween the ages of six and nine (Monsaas, 1985). In contrast, those who werenationally recognized by the age of 12 or 13 began lessons even earlier.Some of these young athletes, such asChris Evert Lloyd, started lessons as early as three years old.

Development of World-Class Tennis Players Monsaas' (1985) report of a study of 18 world-class tennis players,all under 40 years of age, provides a fairly detailed account ofthe development of their interest and ability intennis. Most of the subjects were from middle-class to upper middle-class families. Few of th- mothersworked out...ide the home. Three of these tennis players' fathers were tennisprofes- sionals. In about 80 percent of the families, at least one parentplayed tennis avidly. Nearly all of the subjects spent a large part oftheir time, even before they began to play tennis, at tennis or countryclubs, where they swam and played active gameswith other children. Two qualities of the family lives of these subjects struck theresearcher as unusual: the closeness ofthe family and the intensity with which they worked. Part of the intensity seemed to derive from a spirit ofcompetitive- ness. Just doing well wasoften not considered good enough; being the best at what they did seemed tobe important. The tennis players reported that they considered themselves and were consideredby others to be highly tenacious and competitive. As they became more skilled in tennis, these youngathletes devoted more of their time and energy topractice. Their competitiveness and determination was not so pronounced in their other endeavors asit was in tennis, however. By the time they were ten to 12 yearsold, they were identified, at least locally, as talented tennis players. Theirpractice and playing time had increased, and they were on the tennis courtsabout 20 hours a week or more when weather permitted. Atabout this time, they 2 6;4 Pnreeciosis Devidapand 261 typically changed from their beginning coach to a coach who could help them develop their own particular strengths. The decision to make tennis a career was not usually made until their college years. Most played tennis on college tennis teams and had tennis scholarships. The women players, however, repoi Led that they had difficulty getting scholarships because women's sports were valued less highly than men's sports. All of the world-class tennis players who were interviewed said that they felt pleased with their careers. They felt that the satisfaction of doing something well was the greatest career benefit; but the tennis players also reported that they liked the chance to travel and to earn money doing something that was satisfying.

Development of Olympic Swimmers Kalinowski (1985) studied 21 Olympic swimmers. In many ways, these athletes were similar to the tennis players discussed above. They too showed increasing levels of commitment to their sport as they got older. The swimmers first started their involvement with the sport as a recreational activity; later their training became more intense. Like the families of tennis players, the swimmers' families were close-knit and subscribed to a strong work ethic. The swimmers' families, by contrast to those of the tennis players, usually did noi have an exceptionally strong commitment to their children's chosen sport. The swimmers' initial experiences with their sport were unstructured, another difference between them and the tennis players. Often their first structured involvement was with a team; of those swimmers interviewed, about half had joined a team by the age of seven. The teams practiced about four hours per week, and often the students did not have individual lessons or coaching during that time. Usually swimmers received formal coaching only after they had demonstrated outstanding ability on a team.

OTHER PRECOCITY

Some early talents ar. very specialized. "These talents are so rare and so specific that their relationship to more generalized abilities has not been determined. Talent in three specialized processes has been well documented: eidetic imagery, rapid mental calculation, and chess playing.

Eidetic Imagery (Photographic Memory) Many prodigies are reported to have had photographic memories. Poincare, a famous mathematician of the nineteenth century, could repeat from memory the exact page and line of a quotation from a book after

2'70 262 Precocious Develapsunt reading it once (Bell, 1965). William Sidis, a highly intelligent prodigy, is said to have had a photographic memory as well as rapid calculating ability, "Photographic memory," or "eidetic imagery," is defined as vivid visual imagery of things previously seen. Vividness and complexity of detail distinguish eidetic images from other images of remembered visual phenomenon (Stromeyer, 1970). Eidetic imagery reportedly occurswith greater frequency among prodigies andbrain-damaged children (Paivio, 1971). Sipola and Hayden (1965), however, found no systematic differences in the intelligence of eidetic and non-eidetic chiklren. Paivio (1971) proposed that eidetic imagery might be a functional development encouraged by any factor that restricts development of verbal representation; but, as he cautioned, any theoretical interpretation is highly speculative. Althaugh eidetic imagery is often said to disappearwith the attainment of literacy, this is not always the case (see e.g., Stromeyer, 1970). The degree to which eidetic imagery contributes to outstandingscholarly or artistic performance is alsounclear. Much more research is needed before the relationship between this ability and academic accomplishment can be determined.

Rapid Arithmetic Calculation Smith's (1983) The Great Mental Caktdatars counters some of the popular claims about this type of prodigy. One such claim is that the ability todo lengthy arithmetic operations very rapidly, without pencil and paper, is as often a characteristic of retarded or brain-damaged children as of gilled children. Another is that this ability usually disappears when the child grows up. Although, like other typesof prodigiousness, rapid mental calculation is usually attributed to innate ability, Smith contends that this extraordinary skill is the result of exceptional interest and continual practice. The difference between the development of this skill amongmales and females seems to confirm his hypothesis.Far fewer females than males develop skills as rapid mental calculators. Smith seesthis difference as a function of social values that fail to provide as much encouragementfor those girls whose calculating potential issimilar to diat of boys. According to Smith (1983), there is a distinctionbetween auditory and visual ulculators. The former are usuallyprecocious. They begin rakulat- ing before they begin to read or write. Themathematician, Gauss, began calculating at three; and Ampere between threeand five. Visual calculators, on the otherhand, usually begin in their teens. Like most researchers, Smith contendsthat mental calculating ability does not imply mathematicalability; and there is no reason to think that it is a necessary component ofmathematical precocity. However, such an

271 Piveociess Developmest 263 ability is probably useful for creative work in mathematics. Gauss, for example, said that his work was advanced by a memory that allowed him to work with many numbers at the same time (Bell, 1965). Poincar6's mathematics were done in his head and committed to paper only when a problem had been thought through (Bell, 1965). Galois's "peculiar gift of being able to carry on the most difficult mathematical investigations almost entirely in his head" is described in biographies of the mathematician (Bell, 1965, p. 365). Others who are noted only for their rapid mental calculation might have become extraordinary mathematicians if they had had appropriate educational experiences. Zerah Co lburn (see Box 9-2), for example, never was able to use his ability in the context of the disciplined learning of mathematics.

BOX 9-2 Zerah Coburn: Case Study

Zerah Colbum was born in 1804; he was the fifth son of a Verrnont farmer. When Zerah was five years old, his father overheard him repeating multiplicationtables to himself. When asked to multiply a two-digit number by a two-digit "umber, Zerah could respond immediately with the correct answer. His father started exhibiting Zerah and continued to do so for ten years. However, Zerah was not given a very good education during that period of time. Zerah was in school for only two or three years, and there is no evidence that he was educated at home. At eight, Zerah could factor numbers as large as 171,395 and give the possible combinations of numbers that could be multiplied together to get that product. He could multiply five-digit by five-digit numbers, but six-digit by six-digit problems seemed to be out of his range. Although Scripture's (1891) Arithmetical Prodigies describes Colburn as unable to learn much of anything and lacking in ordinary intelligence, Smith (1983) points out that Colbu'i was, in fact, self-educated and capable in many academic areas. Colburn made calculations related to astronomy, taught modern and classical languages and literature and wrote an autobiography in which he attempted to describe his arithmetic processing. Smith attributes Scripture's claim to snobbery, since Colburn had no scientific and little literary education, Scripture considered him presumptuous in writing about his psychology. Binet also misrepresented Colburn's level of ability. Using only secondary sources, Binet described him as being of mediocre intelligence. Another writer, Jacoby, reported Colbum's first and last names incorrectly and yet said au- thoritatively that "'Zerald' did nothing in college, except calculations from memory; he either did not want to, or could not, le-rn anything" (p. 209). Colbum's supposed ineptitude may be associated with his family background and social status. His father exploited his abilities, whether for his own aggrandizement or out of ambition for his son. An uneducated man himself, he lacked the means to provide his son with a high quality education. Zerah spent an inordinate amount of time practicing and improving his calculating skill, but no commensurate effort toward the development of more critical academic skills. (abstracted from Smith, 1983) 264 Precocious DiPtiopallelti Chess Playing According to de Groot (1978), himselfa famous chess player, expert chess playing does not primarily dependon rapid thinking. Nor does it depend solely on the knowledge of different kinds of chessmoves. Rather it is based primarily on knowledge of and experience withheuristic techniques. This experience is derived froman early devotion to the game. Although there have been a number of chess prodigies,Bobby Fischer was the youngest of these to become internationalgrand master of chess (Brady, 1973). He learned howto play chess when he was six and was invited to play with members of the Brooklyn (N.Y.)Chess Club when he was about eight years old. Throughout his childhood, Fischer practiced chess incessantly and studied from whatevermaster chess players he could find. At 14 he won the U.S. chess championshipfor the first time. Fischer came from a middle-class family. His fatherwas a biophysicist, and his mother was a s -hoolteacher andnurse. Both Fischer's mother and his sister provided him witha sense of emotional security and supported his interest in chess. Althb igh Fischerwas extremely intelligent, he was not willing to devote time or energyto school subjects. His IQ of over 180 and his temperament reportedly made it hard for himto relate to either his teachers or his peers (Brady, 1973). Hewas highly competitive in sports as well as in chess; and his determinationto win was considered one of the reasons for his success. In order to pay more attention to chess, Fischer dropped out of high school at theage of 16. Nevertheless, he was obviously quite literate. He read widely about chess and eventuallywtote articles and a book about the game (see Brady, 1973).

SUMMARY

This chapter examined the development of child prodigies. These highly gifted children, whether identified by virtue of' their extremely highIQ scores or their remarkable talent, are able to perform like gifted adults. Because there are so few prodigies, thereare few research studies that consider their characteristics. This chapter reviewed the researchthat is available and also presented supporting evidence from biographical sources. The discussion considered several kinds of prodigies. It examined their childhood chars :teristics as wellas their adult achievements. The chapter looked at case studies and empirical researchon ( I) children with extremely high IQ scores, (2) mathematical prodigies, (3) musical prodigies, (4) literary prodigies, and (5) world-class athletes. The chapter also considered some types of precocity about which little is known. It reviewed research about the precocious development of specialized skills, suchas eidetic imagery, rapki mental calculation, and extraordinary chess playing. 273 10 Inhibited Development: Underachievement

I. Focusing Questions

11. Etiology of Underachievement A. Psychological Explanations 1. The psychoanalytic hypothesis 2. Personality problems 3.External locus of control B. Sociological Explanations C. Educational Explanations 1. Socializing function of school 2. Teachers' expectations 3. Peer pressures D.Biophysical Explanations Characteristics of Gifted Underachievers A.School Behavior of Underachievers 1. How early does underachievement begin? 2. Whitmore's research 3. Underachievement in high school B. Personality Characteristics of Underachievers 1. Passive aggression 2. Low self-esteem 3. Attribution of success and failure C. Families of Underachieving Gifted Students 1. Parental expectations 2. Other family variables

2 74 265 266 hskibited Devidapwak ihnisrackkoment

IV.Handicapped Gifted Children A.Measurement Problems B.Adult Expectations for the Achievement of Handicapped Children C.Gifted Students with Visual Impairments 1.Visual impaineents and intelligence 2.Giftedness and visual impairments D.Gifted Students with Hearing Impairments 1.Hearing impairments and inteltigence 2.Giftedness and hearing impairments E. Gifted Students with Physical Handicaps 1.Orthopedic impairments and giftedness 2.Hedith impairments and giftedness F.Gifted Students with Learning Disabilities Definitions: brief history and research 2.Discrepancy: brief history and research 3.Characteristics of gifted children with learning disabilities G.Gifted Children wii:i Emotional Problems 1.Deviance as ameasureof emotional imbalance 1.Characteristics of gifted children with emotional problems V.Summary

FocusingQuestions 1. What characteristics of schools might account for the underachievement of some gifted students? 2.How do the personalities of giftedunderachievers differfrom the personalities of gifted achievers? In what ways are these personality factors conditioned by children's home environments? 3. In what ways do adult expectations limit the goals and achievements of gifted students with handicaps? 4.What characteristics of gifted underachievers predispose them to being incorrectly classified as learning disabled? What factors limit researchers' efforts to conduct systematic studies of gifted children with serious emotional problems? 6.How do the minor adjustment problems that gifted students sometimes experience differ from serious emotional disturbance?

ETIOLOGY OF UNDERACHIEVEMENT

Sometimes gifted children's achievement falls short of their potential. The level of jchievement of such children does not match their high level of ability as measured onIQtests. Often, this underachievement is reflected in low grades. This type of underachievement, while frustrating for teach-

2 75 Inhibited Devektfeneon; Undernehireenteitt 267 en, parents, and the children themselves, does notnecessarily indicate a lath of achievement. Some bright children who make poor grades may actually score quite high on standardized tests of achievement. Another, more serious, type of underachievement occurs when agifted child scores significantly below potential on a reliable test of achievement. This type of underachievement sometimes results from a handicapping condition that inhibits the child's development, but, at times, the reasons for this sort of underachievement are more difficult todetermine. About 20 percent of the students identified as gifted earn grade point averages of 3.0 or lower on grade-level work.This fact is surprising when one considers that gifted students usually score atleast two grade levels above grade placement on standardized tests: Gifted kindergarteners score at about the second-grade level, junior high-school students score atthe high school and college level, and high-school seniors score at the level of the average college senior (Marland, 1972). A smaller, but stilt sizable, number of gifted students also score much lower than might be expected on standardized tests of achievement. There are quite a rem hypotheses that attempt to explain the underachievement of some gifted students. Many of these hypotheses account for the school failure of such students, and others address students' discre- pant performances on tests of ability and achievement. Theliterature on underachievement is often difficult to interpret because researchers define achievementand hence, underachievementin various ways. Many studiet. for example, use grades as a measure of achievement (e.g., Dowdall & Colangelo, 1982). Others equate it w:th performance on group achievement tests. The most rigortr.studies tend to characterize it as performance on individually-administered,standardizedtests of achievement (e.g., Mather & Udall, 1985). In addition to these measures of achievement, accomplishments (cf. Rimm & Davis, 1980) and vocational attainment have sometimes been used to study the achievement of gifted adults. Psychologital Explanations Much of the research on underachievement has sought behavioral or attitudinal correlates that distinguish gifted underachievers from gifted achievers. Because gifted underachievers make high scores on tests of intellectual ability, many authors assume that the underachievement of such students has to result from emotionalas opposed to cognitive or school problems. Psychoanalytic theorists hypothesize that unhealthy interactions between parents and children are the most likely cause. Other psychologists attribute the problem to students' personality problems or to an external locus ot zontrol.

The psychoanalytic hypothesis.Psychoanalytic explanations of underachievement view it as a syndrome of conflict within the student that

2 76 Inhibited Devekneeent: Undmehievetneost results from unproductive patterns of family interaction. From this perspective, underachievement is often interpreted as a form of rebellion against parental authority by a child who is afraid to show overt hostility. Psychoanalytic theorists, however, do not seem to agree about the particular types of family interaction that result in children's underachievement. Many writers believe that an unherlthy relationshipbetween the mother and child is associated with underachievement, but authorities disagree about the particular nature of this unproductiverelationship. Some studies have found mothers of female underachievers to be dominant (Fine, 1967); others have found them to be permissive (Raph, Goldberg, & Passow, 1966). Some theorists believe that underachievers (particularly boys) use poor school performance to enact their hostility toward overprotective mothers, but others interpret overprotective maternal behavior as an indication of the mother's hostility toward thechild (e.g., Brick lin & Brick lin, 1967). To make the issue even more complicated, some studies suggest that the problem does not involve the relationshipbetween mother and child at all, but rather involves the relationship between father and child (e.g., Kimball, 1953).

Personality problems.Psychologistswho do not necessarily accept psychoanalytic explanations of underachievementhave found that underachievement is associated with certain personality traits. Terman and Oden (1947), for example, conducted follow-up studies to find out about the school and work success of their gifted sample. They reported that the men who failed to reach theirpotential in school or in their careers exhibited similar personality traits. Parents, teachers, and wives who rated the performance of these men noted that the men (1) lacked the ability to persevere, (2) failed to integrate goals, (3)tended to drift rather than take action, and (4) lacked self-confidence. Parents and teachers had madesimi- lar observations about the childhood personalities of these individuals, a finding that led the researchers to conclude that the underachieving men hadchronicpersonality problems (Tertnan & Oden, 1947). See Chapter 5 for additional analysis of this research. Brick lin and Brick lin (1967) identified the following personality traits of underachievers:

passive aggressivenessbehaviors motivated by the fear of outwarddisplays of anger, identification of self-worth with the ability to achieve. dislike of being considered "ordinary," fear of success, low tolerance for frustration, and the tendency to deal with stress ininfantile ways. (pp. 15-10 277 Inhibited Development: Underachievennent 269

External locus of control.According to some researchers, high and low achievers attribute their success or failure to different factors (Kanoy, Johnson, & Kanoy, 1980). High achievers tend to attribute success to internal factors: ability or effort. They tend to attribute failure to aninternal, unstable factor over which they have some control, that is, lack of effort (Bar-Tal, 1978). Le-cause of this way of viewing their own achievement, these students are said to have an internal locus of control. Low achievers, however, tend to attribute success to externalfactors such as luck or an easy task, but even highly intelligent underachievers often attribute f4ilure to their own lack of ability. These students seem to have anexternallocus of control. As a result, these students deny their competence: They take no credit for their successesand blame their failures on lack of ability. Highachievers work more intently and persevere longer at intellectual tasks. They believe that effort makes adifference. Underachievers apparently do not.

Sociological Explanations Conditions in society may influence gifted children's patterns of achievement and underachievement. Among the most significant of these sociological influences are race, gender, ethnicity, and social class. These factors relate to children's achievement indirectly. They aremediated by factors in the homeparents' education, mother's aspirations,family income and wealth, housingcharacteristics, and parents' occupationthat haveaMOMdirect influence on children's school achievement. Some cases of underachievement may be accountedfor by severe hardship, and a number of studies of underachievementevaluate the effects of such meager environmental support. Insuch environments, a high degree of economic and emotional insecurityprecludes the involvement of parents in theirchildren's educational efforts. This etiology, however, does not account for theundei achievement of children from affluent homes in which education appears to be valued. Sociological variables are such powerful influences onstudents' achievement that they need to be considered in somedetail. Therefore, the discussion of the effects of race, gender, ethnicity,geography, and social class on gifted children's achievement willhereserved for Chapter 11.

Educational Explanations Except for the family, the school may be the mostpowerful social institution in modern society. In part. the influence of theschool relates to its role as "culture-bearer," a role that empowersthe school to convey the most important values of ourculture. Teachers often present these values overtly when they state expectations forbehavior, but they also teach sonic 278 270 inhibited Developmesk Underachievement values through their covert actions. Teachers' differential treatment of children may convey the degree to which they endorse particular characteristics or behaviors. Although individual teachers may vary in their values, some values appear to be very common among teachers. Inaddition to conveying these values, the school also perpetuates the values of the peer- group culture that it houses. Three perspectives on education explain the school's influence on the achievement and underachievement of students: (1) explanations that relate to schools' socializing function, (2) explanations that relate to teachers' expectations, both overt and covert, and (3) explanations that relate to the expectations of the peer group.

Socializing function of school.Holt (Dulee, 1983) has summarized the socialization functions of schools quite well. He notes three basic socializing functions: (1) custody, (2) ranking, especially as an internalized sense of success or failure, and (3) compliance.These functions have been touched on in various contexts by many educational writers (e.g., Baer & Bushell, 1981; Bloom, 1977; Coleman, 1961; Cremin, 1961; Cuban, 1982; Jencks et al., 1979; Katz, 1971; Kohl, 1967; Raskin, 1972). Through these three functions, schools convey the fundamental values of society. These functions also enable schools to socialize children to adult work roles. Schools' emphasis on conformity reflects the priority that our society places on affective as opposed to cognitive performance. It also reflects the prevalent anti-intellectualism in our culture. Many classrooms are unstimulating and confining. A study reported by Marland (1972) found that in over half the regular classrooms, students were not just neutral, but were negative and uninterested. Emphasis in most regular classes is on neatness, recall of factual information, and compliance. There is little room for reflection or for divergent thinking and behavior. Even very young gifted children are dissatisfied with school because of pressures to conform, unfiiirness of teachers, and an irrelevant, unaim- ulating curriculum (French & Cardon, 1968). Gifted underachievers from the first, second, and third grades give the following reasons for their discontent: resentment or hostility from teachers and peers, social penalties for nonconformity, lack of opportunities to pursue individual interests, teacher criticism, teacher control, and an unrewarding curriculum (Whit- more, 1980, pp. 192-193). Of these variables, theunrewarding curriculum is the most damaging to gifted children as a group. These children are constrained by educators' expectations of what children of a certain age can learn. All of the classroom variables listed by these young underachievers are related, however. They aremanifestations of the essential problem facing such children: pressure to conform. Personality and attitudinal traits common among gifted underachieversactually make them more znanerabk

27!4 Inhibited Development Litsdentichseve- mesa 271 than other bright children to pressums for conformity (Whitmore, 1980). Achievers appear to have the ability to protect themselves fromsuch pressures; they may appear to conform,but nevertheless base their choices on personal values. Underachievers seem less able to tolerate such ambiguity.

Teachers' expectations.Teachers verbalize many of their expectations for children's behavior. Most often these expectations relate tochildren's affective rather than cognitive behaviors. According to Baerand Bushell (1981, p. 264): .. recentobservational research suggests that teachers invest most of their behavior not in teaching academics but in maintaining order .. .and that teachers prize and acknowledge conformity with these rules moreoften and more vigorously than they do correctacademic problem solutions. . The sane research suggests that students learnexactly that emphasis from teachers and believe that it is much moreimportant to follow the rules and be good citizens than to solve problems correcdv, [Baer and Bushell's emphasis) Nonetheless, teachers' expectations are not always overt. Recent research shows that expectations are conveyed through teachers' actions in the classroom. Usually, these actions reflect social stereotypes. Teachers favor children who are attractive, neatly dressed, middle class, high achie.v- ing, and conforming. They respond less positively to children who are culturally different, economically disadvantaged, low achieving, and unat- tractive. Curiously, teachers also respond negatively to divergent and highly intelligent children. In the classroom, teachers' biases are evident in their reactions to certain children. They call more often on the students whom they like; they give more positive feedback to these students; they seat them in more accessible locations in the classroom; and they provide them with more appropriate work. (See Good & Brophy,1987, for an excellent summary of the research on teachers* expectations.) It is clear that stereotypes affect teachers' behavior toward different sorts of students. An example is provided in a study that comparesteach- ers' directives in a working-class school, in a middle-class school, in an affluent school, and in a highly elite school (Anyon, 1980/1987). Similar results are reported in a study by Wilcox (1982). The teachers in the affluent and highly elite schools encouraged students to internalize academic discipline. The following examples of their verbal messages to students reveal their expectations: Decide what you think the best way is. Don't be afraid to disagree. In the last Imathl class, somebody disagreed, and they were right. Even if you don't know (the answersj, if you think logically about it, you can figure it out. I'm asking you these questions to help you think it through. (Anyon. 1980/1987, pp. 220-223) 250 272 lodtibited Devekpeeent: Underaehievelsest These teachers explained the reasons for their actions and required the students to be responsible for their own academic progress. The teacher in the working-class school, on the other hand, emphasized external control. Her instructions for students' academic work and behavior induded the following statements:

Remember when you do this, it's the same steps over and ovet again and that's the way division always is. Throw your gum awayif you want to rot your teeth, do it on your own time. Do it this way, or it's wrong. (Anyon, 1980/1987, pp. 214-215)

The differences in teacher directives in Anyon's study reflect differences in the types of adult roles the children are expected to fill. Upper middle-class children are more likely to assume professional roles that entail self-direction. Roles that the working-class children are likely to fill involve much closer supervision. According to Anyon (1980/1987):

School experience ...differed qualitatively by social class. Differing curricular, pedagogical, and pupil evaluation practices emphasize different cognitive and behavioral skills in each social setting and thus contribute to the develop. mem in the children of certain potential relationships to physical and symbolic capital, to authority, and to the process of work. (p. 225)

Directives to students are one way that teachers convey expectations based on social stereotypes, but grades also reflect teachers' prejudices. Often they are assigned on the basis of students' conformity to behavioral expectations rather than on the basis of academic achievement (Baer & Bushell, 1981; Bloom, 1977; Glidewell Kantor, Smith, & Stringer, 1966; Kubiszyn & Borich, 1984). Gifted students, who see themselves as disorderly, opinionated, and fault-finding (Pyryt, 1979), are at risk with respect to the expectations held by schools. Someregardless of their aptitudewill refuse to comply with these expectations and will, therefore, fail to earn good grades. Others will avoid academic work to such a degree that their achievement (on standardized tests) suffers; and still others may never identify with intellectual values at all.

Peer pressures.ln schools, the peer group establishes expectations for behavior. These expectations are as limiting as teachers' expectations. The peer group often includes a number of cliques, the most prestigious of which base membership on the class background, race, and personal characteristics of students. Poor children, minority children, !ow achievers, and extremely high achievers are often excluded from membership. Coleman's (1961) study, Adokscent Society, characterized the values of high-school peer groups. In his study Coleman identified three status

2S1 Inhibited Development: Undmuchieventam 273 categoriesbest scholar, best athlete, and most attractive student. Infor- mation from Coleman's study allows us to compare the position of attractive individuals, top athletes, and top scholars within the peer-group culture. Top athletes were generally members of a top clique; top scholars were more frequently not clique members at all. Of the 32 top scholars identified in the study, half the girls and about three-fourths of the boys were not identified as members of any clique. Those who did belong to cliques were found at the fringes of low-status cliques. There were four isolates (i.e., students who had no mutual friends) among the 96 top-status students in the study. Three of the isolates were top scholar girls. Seldom were the top scholar and the top athlete of a class the same person.The combination was equally rare for both sexes. No top scholar, of either sex, was also selected as most attractive, but many top athletes were selected as most attractive. Considering Coleman's findings, it seems likely that adolescent peer groups actually discourage the scholarly pursuits of some students. Capa- ble students who are not willing to give up their academic interests risk rejection by their classmates. Such powerful influences probably contribute to some gifted students' underachievement, though the nature anddegree of their influence might be difficult to discern.

Biophysical Explanations Sometimes, underachievement results from students' physical. neurological, or emotional problems. Bright children who are handicupped ofter perform like other gifted underachievers: They may have difficulty getting good grades in regular classrooms. However, unlike some other underachievers, handicapped gifted students may also demonstrate discre- pant test performance. Their IQ scores may far exceed theirachievement test scores. The low achievement or distortedbehavior of these students may bring them to the attention of their teachers. In some cases, gifted children's handicaps may alter achievement or behavior only enough to mask students' giftedness. Some gifted children use their exceptional ability to compensate for theirdisabilities. Because their compensation strategies are reasonably successful, such children may appear normal to their teachers. Whenstudents compensate for their disabilities in this way, schools are likely to overlook both their giftedness and their handicaps. There are many blind, deaf, and physically handicapped children who are also gifted. Fortunately, such students are not always underachievers. Sometimes they make successful adaptations that allow them to minimize the disibling effects of their impairments. Special educational services often help them achieve at levels commensurate with their abilities. Nevertheless, in many instances these children do not achieve as well as their intelligence test scores suggest that they should.

2 8 2 274 Inhibited Derelopors4- thodarachievened Other handicapped children, those referred to as specific warning disabled (SLD) and emotionally disturbed (ED), are characterized by their underachievement. Part of the process of identifying students with these handicaps involves the documentation of their academic underachievement (Cartwright, Cartwright & Ward, 1981; Lewis & Door lag, 1983). These students, like other handicapped gifted students, often encounter serious difficulties in school. Whitmore (1981, pp. 109-110) considerssey- Pral of these difficulties, including; expectations basedon preconceptions about the handicapping condition, developmental delays that obscuresu- perior performance, and lack of opportunities to demonstrate exceptional mental abilities. Some writers consider specific learning disabilities and emotional disturbance to be biophysical causes of underachievement (see e.g., Johnson & Myklebust, 1967; Ritvo, 1977). These theorists believe that biochemical and neurological abnormalities account for the learning and behavior problems of such students. Studies of the brain chemistry of schizophrenics and research into the neurological basis of anentional deficits are examples of this approach. Some theorists, however, do not characterize specific learning disabilities and emotional disturbance among the biophysical causes of underachievement. These writers emphasize the role that school and society play in causing children's learning disabilities and emotional problems (seee.g., Carrier, 1983). According to one authority (Whitmore, 1982), the emotional problems and learning disabilities of gifted children may be caused by social limitations on their ability to progress academically, expressemo- tions, or pursue interests.

CHARACTERISTICS OF GIFTED UNDERACHIEVERS

This section of the chapter reviews the empirical research on the characteristics of gifted underachievers whose difficulties result from a cause thatis not biophysical. As indicated in the previous section, different theories attempt to es-plain this kind of underachievement. Since none of these theories is conclusive, however, our discussion is not based on any single theory. Instead, it examines this kind of underachievement topically, discussing school behavior, personality characteristics, and families.

School Behavior of Underachievers A number of researchers have assessed the school behaviors of underachieving students. This research suggests that, in general, underachievers dislike school and do not comply with teachers' or parents' expectations for 2s 3 Inhibited Dovekipseist: Undenschismissot 275 their performance. According to some research, students' patternsof underachievement begin early and get worse as they progress through the grades.

How early does underachievement begin?Convinced that underachievement results from students' maladaptive responses to problems. Shaw and McCuen (1960) tried to discover when underachievementbe- gins. They selected a sample of students whose 1Qs placed them inthe upper quartile of the population.Based on the students' grade point averages (GPAs) in the ninth, tenth, andeleventh grades, the researchers divided the sample into four groups: male and female achievers and maleand female underachievers. Then the researchers reviewed these students' school records to find out how they performed at each grade level. As a result of their research, Shaw and McCuen (1960)identified distinct patterns in the performance of underachieving males andfemales. Beginning in the third grade, the grade point averages of the maleunder- achievers were significantly lower than the GPAs of the male achievers. The female underachievers, however, showed a different patternof performance in the early grades. From the first through the fifth grades,these students actually made better grades than the female achievers.When they were in the sixth grade, theachieving females received higher GPAs than the underachieving females for the first time. From thesixth until tenth grade, this difference increased, al*hough it was notstatistically significant until the ninth grade. In spite of the suprising findings for femaleunderachievers, the authors of this study concluded that underachievement was achronic problem for both sexes. They suggested that both the malesand the females entered school with predispositions that kept them fromsucceeding.

Whitmore's research.Whitmore (1980, p. 79) described three behavioral syndromes of underachievement: (1)the aggressive, (2) the withdrawn, and (3) the erratic. She compared thestudents in her group to find out the characteristics mostcommonly associated with these types of underachieve ment. She also identified characteristicsthat were common to all of the gifted underachievers in her sample. Whitmore (1980) reported that all of theunderachievers that she studied had the following characteristics:

very high iQsI40 or above on eitherthe Stanford-Binet or the WISC-R, tendency not to complete assigned school work, vast difference in the quality of oral andwritten work, school and test phobia, 2S4 276 Inhibited Development Undsraehieventent

strong interest in one specialty area, very low self-esteem, sincere belief that they are not liked by others, autonomous spirit, inability to work well in a group of any size. tendency to set goals that are ioo high, dislike of repetitive tasks, and failure to respond to typical classroom motivators (e.g., praise or awards).

Less common, but still found frequently, were characteristics such as dis- tractibility, hyperactivity, immaturity, psychomotor inefficiency, lack of academic initiative, chronic inattentiveness, and hypochondria. According to Whitmore (1980), approximately 75 percent of the underachievers that she studied were of the aggressive type. These students often refused to comply with rules. They were disruptive in the clissroom and sought attention in inappropriate ways. These students seemed to lack self-direction and self-control. They attempted to find any opportunity for avoiding assignments. They also had difficulty getting along with peers and tended to be either physically or verbally abusive toward them. Whitmore noted that most of the aggressive underachievers in her group were boys. She related this phenomenon to cultural expectations for boys' behavior. " smaller proportion of underachievers were withdrawn. They rarely communicated with peers or with the teacher, and they tended to daydream. These students did not work well either in groups or individually. Often, they ignored assignments. When attacked by more aggressive children, the withdrawn underachievers failed to defend themselves. A few of the students in Whittnore's study vacillated between aggressive and withdrawn behaviors. She labeled these children "erratic." Accord- ing to Whitmore, students who were classified as erratic were more likely than the other students to have some identifiable biophysical ba.7:s for their underachievement.

Uaderachievensent in high school.Identifying underachievement among high-school students is more problematic thanidentifying underachievement among elementary students. Some high-school students who make adequate grades are actually achieving considerably below their potential. Such discrepancies are difficult to measure, in part because the popular individual achievement tests (e.g., PIAT, Woodcock-Johnson) do not have sufficiently high ceilings for older giftedchildren. Another group of high-school students make poor grades in spite of high achievement test scores. These students fail to comply with the mum- cognitive expectations of school. Their behavior is understandable in light of the fact that the high-school curriculum may be inadequate for them. assignments may be trivial or unchallenging, and teachers may be unre- 285 hshibited Develop/Nod: thedenickievezent e77 sponsive to their needs. Unfortunately, such students suffer in the long run from their lack of attention to school assignments. Although some of these students can tio well as undergraduates, many of them lack the necessary habits of scholarship to succeed in (or enjoy) college classes. Students who behave as if scholarship were not a serious matter, or as if effort were irrelevant to learning, are at risk academically, even if their high-school achievement test scores are not bad.

Personality Characteristics of Underachievers Much of the research about underachievers considers the personality characteristics of such students. This research should be interpreted cautiously, however. Because of the nature of their school problems, most underachievers experience a great deal of frustration. The behaviors and attitudes that are presumed v iacterize underachievers may, in fact, reflect underachievers' high le f frustration. In more supportive environments, underachievers might show quite different aspects of their personalities.

Passive aggression.The conclusion tIrat underachievement is a means of expressing hostility toward an overly demanding parent recurs throughout the literature on underachievement. An early study of the personality development of underachieving college students described these students' explanations for their academic failure as "unrealistic, su- perficial, and largely implausible" (Kirk, 1952. p. 214). Kirk viewed underachievers' excuses as an effort to control their hostility; she believed that the academic failure of such students might be an unconscious manifestation of their hostility. Similar condusit ns seem to have been supported by a study of underachievers in high school. Shaw and Grubb (1958) reported that, on three of four personality scales, underachieving males had hostility scores that were significantly higher than those of achieving males. However, the study found no significant difference between the scores of underachieving and achieving females. Because the scale did not relate to hostility toward school, the researchers concluded that the genesis of the problem had to be elsewhere. They determined that the problem was likely to be caused by dynamics in the family. Another study of high school students used sentence completion items as a projective test of personality characteristics (Kimball, 1952). This study found that underachievers were less likely than achievers to complete the sentences in ways that suggested the outward display of aggression. Even in situations where aggressive action would be appropriate, many of these students failed to give aggressive responses. By contrast, many more of the achievers gave aggressive responses to such items. 2S 6 278 Inhibited Development thdorachirreumnd Roth and Puri (1967) also used projective techniques to assess the degree of outward aggressiveness exhibited by underachievers. Their study, like the others reported here, concluded that underachievers either turn their hostility inward or repress it rather than expressing it openly. These researchers asked students to respond to a picture of a scene in which aggressive action was required in order to solve a problem. Rather than suggest aggressive action, however, the underachieving students tended to verbalize intrapunitive (inwardly aggressive) action. Their responses were contrasted to those of achievers whose responses favored extrapuni- tive (outwardly aggressive) action. The responses of some underachievers involved no punitive action at all.

Low sev-eskem.Quite a few studies have explored the relationship between self-et;teem and achievement. In general, achievers have more positive self-concepts and higher self-estecm than underachievers (Whit- more, 1980). However, it is difficult to tell whether lowself-esteem is a cause or a result of poor achievement. A study by Fink (1962) compared psychologists' ratings of achievers with their ratings of underachievers. This study showed that among the boys low achievement was associated with low self-esteem, but this finding did rm. l hold true for the girls. In other studies (e.g.. Mitchell, 1959), however, underachieving girls seemed to be more like the boys in Fink's study: They were more self-rejecting than achieving girls. Although many studies yield similar results, Raph and associates (1966) caution against uncritical acceptance of this research. Studies of self- concept are limited because of confusion in both theory and practice. The- orists conceive of self-concept in different ways; and researchers do not agree on what tests to use in order measure students' self-esteem orself- concept. Related characteristics, such as self-image, self-perception, self- acceptance, and self-confidence are sometimes equated withself-concept. Until there is agreement about the nature of self-concept and its quantifica- tion, research results that link underachievement to a negative self-concept should be viewed quite circumspectly.

Arnibution of success and failure.Many underachievers attribute their failure to poor teachers, bad luck, lack of ability, or an unstimulating curriculum (Gallagher, 1985). Bright students who do poorly in school seem to feel that they cannot improve their performancebecause they lack control over the factors that influence academic success. According to Bar-Tal (1978), high and low achiever.; differ in the reasons that they give for their successes and failures.This research grouped the reasons that students gave for their success or failure on an academic task into four categories: ability, effort, luck, and task difficulty (Bar-Tal, 1978). High achievers usually attributed their failure to lack of

2S7 Inhibited Devekpand: Underadnevente' 279 effort, whereas low achievers attributed their failure to lack of ability. Low achievers attributed theirsuccess,however, to good luck or an easy task. These students denied their own competence and refused any credit for success. Discounting their own agency, they seemed to see themselves as powerless victims of chance. Davis and Connell (1985) reported different results. These researchers discovered that the underachievers in their sample of 122 gifted and average students ascribed failure to lack of effort,notto bad luck or a difficult task. They also reported that the underachieving children were lesslikely than achieving students to credit the influence of teachers on their academic performance. That is, in this study, the underachieving children demonstrated agreaterinternal locus of control than the achieving students. This is a surprising result. lt contradicts the findings of other studies about underachieving students. Perhaps Davis and Connell's results stem from the design of the study. which examined the comparative influence of achievement on a wide variety of self-related constructs. On the other Irind, the unexpected results might also be explained by the nature of the sample studied. All the students apparently came from middle-class homes. As noted previously in this chapter, schools cultivate an internal locus of control among middle-class children. It is possible that attribution of success and failure varies by social class, though this hypothesis has not been investigated among gifted students.

Families of Underachieving Gifted Students Pringle (1970) blamed the home environment for the school difficulties of mmt underachievers. She nottd that in the families of under- ach:f!vers at least one parent tended to be either too indulgent or too strict (Pringle, 1970). A number of other researchers have reported similar results.

Parental expectations.Too little parental demand for achievement may leave children without incentive to persevere on arduous tasks (Fine, 1967; Pringle, 1970). Too great a demand may 'aring resistance. The underachievers' negative response to authority is often regarded as rebellion against overly demanding parents. Afraid to express their anger openly, they simply fail to perform, often attributing their poor performance to factors beyond their control, for example, that they are not very bright (Raph et al., 1966). Some studies suggest that the parents of underachievers set overly high standards for the performance of their children. Similar findings, however, have been reported about the parents ofachievers.For example, one longitudinal study of bright children claimed that the parents of high

2 S 8 280 lnhii.ited Development: Underachievement achievers were too pressuring (Haggard, 1957). In fact, the high achievers in this study exhibited more resentment toward their parents and other adults than did the underachievers. Is pressure to achieve harmful to gifted children? The answer depends to a large degree on the sort of support that such children get. In order to live up to their potential, gifted children must be expecttd to perform at advanced levels; but they must also have the instruction and support necessar) to accomplish difficult tasks. Children can respond to expectations for high-level performance only if they have the inherent capabilities and the environmental support required for high achievement. If they do not, they may become frustrated.

Other family variables.The discipline styles in l'amilies of achievers and underachievers do not appear to differ in any predictable way. Al- though many underachievers report that their fathers are unsupportive (Pringle, 1970), many achieving males also make sir.-ilar assertions about their fathers (McCle1la,x1 et al., 1953). Results that link underachievement to maternal overprotectiveness are also inconclusive (Bricklin & Bricklin, 1967). Raph and associates (1966) identified no differences in the maritA harmony, protectiveness, regularity of home routine, or pressure for achievement atnong parents of male achievers and underachievers, f low- ever, they did find differences in the degree to which parents shared activities and ideas with their children. The parents of achievers appeared to be more involved with their children's development. 'I'hey were more likely than the parents of underachievers to provide their children with the resources and encouragement necessary for academic success. Pringle (1970) also found that despite above-average sociommomic and educational levels, parents of bright underachievers failed to provide their children with stitnulating cultural anti social oppornmities.

HANDICAPPED GIFTED CHILDREN

The incidence of giftedness among handicapped children and adults is not known. There may be as great an incidence in this population as in the nonhandicapped population generally. However, given the t oncu .rence of mental impairment and severe handicaps, the incidence of gihedness among the handicapped may be less than it is among the nonhandicapped. Regardless of the exact degree of incidence, there is evidence of a large population of unidentifird gifted children among those identified as handicapped. The purpose of this section is to discuss the relationship between various handicapping conditions and giftedness.

2 Inhibited Development: Underachievement 281

Measurement Problems Sometimes educators have a difficuh time measuring the giftedness of handicapped children. In the case of children with visual, auditory, and physical impairments, the usual tests and test administration procedures may not be appropriate. Forexample. evaluating the IQ of a child who cannot hear on a test like the W1SC-R, winch was normed on apopulation of children who can hear, is lxith unfair and misleading. Such children should be compared to other hearing impaired students or should be evaluated with tests designed specifically for the hearing impaired. Less obviovs, though perhaps as troublesome, are the problems en- countered when attempting to measure the intelligence of learning disabled and emotionally disturbed children. Certain learning disabilities, fOr example, are known to interfere with verbal processing. When intelligence is equated with verbal reasoning, the abilities of such children will probably be underestimated. This difficulty is compounded because there are w IQ tests normed on a population of learning disabled or emotionallydisturbed children. When the nature of the child's problems is known, however, evaluators can choose an IQ test that includes more items that will allow the student to demonstrate abilities, and fewer items that will reflect the student's disabilities. Adult Expectations for the Achievement of Handicapped Children Both teachers and parents often have lower expemnions fOr handicapped than for nonhandicapped children. These expectations can have two detrimental effects: ( I) they limit the goals and achievements ofhand- icapped children and (2) they narrow the curriculum made available to such children. Adults who hold low expectations fin the performance of handicapped children often fail to provide them with learning experienues that are suffi-iently challenging. Examples of thedamage caused by lowered expectations were reported in a retrospective study of' handicapped scientists (Maker, Redden, Tonelson. & Howell, 1978). This study identified some of the problems that the scientists experienced duringtheir years in school. Whitmore and Maker (1985, pp. 244'245) pi ()vide a list of the behaviors that reflect adults' lowered expertations for the performance of handicapped children. Low expectations are indicated by tlw f011owing adult reslxnises:

.(rectaig; conversation or giving instructions to others rather than to the disabled perstm: accepting one failure a7, proof that a task cannot he accomplished:

2:m 282 Inhibited Development: Underachievement

giving higher grades relative to others; accepting or giving excessive praise for inferior work; using simple words and sentences with individuals who have cerebral palsy or a hearing impairment because their speech isdifficult to understand; attempting to put a disabled person into a vocational program, despite the person's eligibility and interest in a college preparatory program; excluding a student with cerebral palsy from the chalkboard because the student has difficulty walking.

The ill-effects of such insidious discrimination are frightening. They may limit the achievement of brighthandicapped children, and they often limit such students' access to appropriate programs and services.

Gifted Students with Visual Impairments Visual impairments are usually defined in term, of acuity (i.e., sharp- ness of focus) or field of vision. A visual acuity scoreof 20/70 or less, with correction, is consid-red to represent a visual impairment; 20/200 or less, with correction, is considered legal blindness. Extremely limited field of vision, typically a visual arc of 20 degrees or less in the best rye, is also considered a visual impairment. These arbitrary cutoffs are used to determine which students are eligible for special instruction, special materials, and vision aids. Many visually impaired Audents have enough vision LO use large print books and other visual materials, and even students who arc legally blind may have enough functional vision to use such materials. For educational purposes, those students categorized as functionally blind (as opposed to "legally blind") are those who must rely primarily on braille and other nonvisua! materials.

Visual impairments and intelligence.Visual impairments air caused by disease, accident, and congenital conditions, both hereditary and acquired. Usually visual impairment is not directly associated with mental deficiency. In some cases, however, prenatal viral infections cause borh visual and mental impairment. The factors that cause cerebral palsy may also affect Ixrth vision and intelligence. Visual impairment itself may influence learning and, consequently, affectintelligence and achievement test scores. Some studies have found the average IQ score of blind and visually impaired children to be slightly lower than that of ,rther children. Caton (1985), however, interprets these findings cautiously. She explains the slight difference as a function of testing diffit uhies. She concludes that, "the intellectual development of children is not directly affected by visual loss" (Caton, 1985, p. 255). On the other hand, the academic achievement level of visually impaired children is significantly lower than that of other children. Often it is

291 Inhibited Devekfrment: Undetachirvesient 283 nearly two years behind that of other children (3irch,Tisdall, Peabody, & Sterret, 1966; Caton & Rankin, 1980).The effect seems to depend primarily on the degree of the impairment and the ageof its onset. However, it is difficult to est:trate the effect because ofseveral confounding variables: (1) testing pnx .-Itires are ofteninappropriate, (2) many blind children enter school at attage than other children,and (3) most blind children are educated in trarate schools. Special schools mayaffect achievement test scores because they have a slightlydifferent curriculum than regular schools and a somewhat slower paceof instruction. Addi- tionally. blind children do not have theopportunities for visual experiences, including reading,that promote the rapid acquisition of concepts. Visual impairments have been shown to have a morenoticeable effect on specific types ofintellectual performance than on general intellectual ability. For example, visual impairment seems to affectchildren's ability to asstwiate ideas with objects (Kephart, Kephart, &Schwartz, 1974). Con- cepts pertaining to spatial relationshipsalso pose partkular difficuhies for visually impaired children.

Giftedness and visual impairments. Some subjects, such asmathema- tics and music, seem to give visually impaired children lesstrouble. In tilos*. subj,xls, visually impairt I . tudents may demonstrate their ability by rapid mastery of the curriculum. Suchchildren are not, by virtue of their impairment, predisposed to perfOrm more competentlyin these fields, however. According to Whitmore and Maker (1985, pp. 83-84),indicators of giftedness are the same for children with visual impairmentsand those with normal vision. These indicators include "exceptional memory,advanced problem-solving skills, superior verbal communication,and creative production. .. However, educators will probably have difficulty recognizing these indicators it visually impaired children'sachievement and intelligence tests scores are depressed.

Gifted Students with Hearing Impairments As with visual impairments, hearing impairments canbe classified on the basis of their severity. Hearing acuity is usually measuredwith reference to decibels, or the loudnessof sounds, and pitch. or frequency levels of particular sounds. Persons who show a 20 to 40 decibelhearing loss are usually considered mildly impaired. Those whose loss isfrom 40 to 60 decibels are considered to be moderately impaired. Losses greater than60 decibels are classified as either severe or profinind depending ontheir degree. Because of sensorineural damage, some persons experience aselec- tive loss of hearing. These individuals can hear sounds in somefrequencies but not in others. The most common selective loss affects hearingof high-

29 284 Inhibited Development: Underachiroemmit frequency sounds. This type of loss often affects individuals' ability to hear consonants, the speech sounds which have the highest frequencies. If undetected, this type of hearing loss can prove to be particularly troublesome. Hearing impaired persons may misunderstand words, because of the inability to hear plural or past tense endings. They may respond inap- propriately and appear unintesiigent. In fact, it is not uncommon for persons with undetected hearing losses to be classified as retarded(Whitmore 8c Maker, 1985).

Hearing impairments and intelligence.Verbal intelligenceis so closely associated with language that it is nearly impossible to separate the two. For this reason, children with hearing impairments tend to score lower than other children on verbal tests of intelligence (Green, 1985). Such tests can be administered using an interpreter to sign the questions anddirec- tions, of course. Nevertheless, the language basis of most verbal tests is considerable, and sign language interpretation may not be sufficient compensation for hearing impaired childrens' lack of exposure to oral language. Therefore, childrm who have had a severe hearing impairment since they were very young can almost never perform so well as children with normal hearing on verbal intelligence tests. Those studies that have used nonverbal tests, however, have found little difference between the average intelligence of hearing impaired children and that of children with normal hearing (McConnell, 1973). Children with severe and profound hearing impairments sometimes have serious language problems. Even when their lip-reading skills are good enough to allow comprehension, they have difficulty responding to what is said. Because they cannot bear the way words are articulated, their speech is often labored or distorted ((;reen, 1985). Their speech problems sometimes cause them to be judged, by uninformed listeners, as unintelligent. The academic skill most affected by a severe hearing loss is reading. Since reading is used in so much of school learning, the reading deficits of hearing impaired children help to account for achievement levels that are several years below those of other children their age (Whitmore 8c Maker, 1985). The reading difficulties of hearing impaired students also contribute to others' mispercepfions of their intelligence. Since their written language is also likely to oe affected by their limited language experience, hearing impaired students may alsn fail to express ideas clearly in writing. This, too, has a negative impact on others' perceptions of their abilities. The lack of experience with oral language limits the vocabulary and grammar of many hearing impaired children. Its effect im concept development also seems to be mnsiderable, although research studies have not determined the nature and degree of this influence. We know, however, that to learn most verbal concepts, children need to understand other,

2 3t Inhibited Development: Underachievement 285 more rudimentary verbal ctnwepts. Lackingthese fundamental verbal concepts, hearing impaired children often cannot form morecomplex ones. It is not surprising that, as hearing impaired children get older, their achievement lags further and further behind that of childrenwith normal hearing (Whitmore & Maker, 1985).

Giftedness and hearing impairments.Ilearing impaired children who are identified as gifted are likely to be either extraordMarily intelligent or only mildly hearing impaired. Some hearingimpaired children's aptitude may be apparent only in their performance of nonverbal tasks anti in those academic disciplinessuch as mathematics or dwmistrythat are not highly dependent on language. One case study describes the characteristics of an exceptionalMdivid- ual who had been hearing impaired since his infancy or early childhood (Whitmore & Maker, 1985). In spite of his impairment. he managed to function so well that his profound hearing loss was not detected until he was in the second grade. This child apparently learnedhow to lip-read without finmal instruction lie also learned to read on his own, and, like most gifted children, be read well aii s! enjoyed reading.Iiis at hievement school was high, and he earned a "B" average in both college and medical schtnri.

Gifted Students with Physical Handicaps There are two major categories of physical handicaps: orthopedic impairments and health impairments. Orthopedic impairments include nmscular anti skeletal problems caused by accident. disease, heredity, anti prenatal or perinatal complications. Cerebral palsy, spina bifida. muscular dystrophy, .juvenile rheumatoid arthritis, paralysis, and the absence of limbs are types of orthopedic. impairments. I ieahh impairments include diseases. allym gies, and oilier debilitating physical conditions that do not affect the musculoskeletal 1.vstem. Diabrtes, asthma, heart defects, and epi- lepsy arc aimmg the most common severe health impairments in children.

Orthopedic impairmcrts and giftedness.Some orthopedic impairments are associated with mental impairments.Cerebral palsy, fin L'Xa11)- pie, which may affect only motor control, can also al fect mental development. Approximately 70 percent of the people who havecerebral palsy also have a speech disorder of at least a mild degree; and approximately 50 percent have I(2s below 70 (Cross, 1985). It should berecognired that IQ scores for these individuals are often inaccuraterepresentations of their abilities. Spina bifida, too, is sometimes accompanied by mental roarda- tion. In some instances, however-. spina bifida is associated with precorions verbal developnwnt (Cartwright et al., 1981). in general. the incideme of 29 4 286 Inhibited Development Underachievement giftedness among the orthopedically impaired is the same as in the nonhandicapped population.

Health impairments and giftedness.Very few health impairments affect intellectual development. Ilowever, such impairments may obscure children's abilities. For example, frequent absences due to illness may lower the academic achievement of health-impaired children. In some instances, the medication prescribed for these children may cause them to respond more slowly and to comprehend less well. Parents and teachers may fail to recognize such children's true abilities. Gifted Students with Learning Disabilities In order to understand the characteristics of gifted children who are learning disabled (1.1)), it is important to know something about this handicapping condition. Unfortunately, research about learning disabilities is difficult to interpret. No consensus has been reached concerning the definition of I.D or the cause of this disability. According to Kirk and Gal- lagher (1983. p. 366), "the label 'learning disabihty' includes the heterogeneous group of children who do not fit neatly into the traditional categories of handicapped children." Because there is no accepted definition of I.D, educators cannot agree on the best way to identify 1-11 children. Many presume that learning disabilities origMate in perceptual or conceptual deficiencies. but such causes cannot be measured directly with much accuracy. "The cause of a student's learning disability thus remains an untestable hypothesis. As a result, the diagnosis of the learning disability is most often made on the basis of inference from a severe discrepancy between the student's intelligence test score and achievement test scores. The discussion below provides a lwiel review of the history and research about two crucial issues related to learning disabilities: definition and (iscrepancy determination. It provides a basis for understanding the next section, which considers the characteristics of' gifted children with learning disabilities. Definitions: brief history and research.The definition of learning disabilities guides educators' ef forts to identify IA) children. When there is no consensus about the definition, however, certain problems arise. First is the problem of identifying too many students as learning disabled. During the decade be.ween 1970 and 1980, more and more students were identified as I.D: The proportion of all students labeled I.D grew from 1.3 percent in 1970 to 3.4 percent in 1982 (Smith, 1983). By the 1985-1986 school year 42.8 percent of all handirapped students were classified as 1.1) (flume, 1987). This magnitude of increase had not been anticipated. Mac- intosh and Dunn (1973), for example. indicated that among all students a 295 inlaited Development Undernehieveman 287 prevalence of 1 percent to 2 percent was to be expected.Though writers like MacIntosh and Dunn attempted to drawdistinctions that were sufficiently careful to prevent overidemifkation, their efforts were notuniver- sally effective. The issue of definititm is problematic alsobecause too many different kinds of students have been said to be LI)(Algozzine. Ysseldyke. 14-. Shinn. 1982). The present confusion derives in partfrom the influence of a variety of historical constructs. forexample "dyslexia" (Orton. 1937); "minimal brain injury" (Strauss & Lehtinen, 19,17); and"psychoneurological learning disorders" (Mvklebust & Bushes, 1960).Bryan and Bryan (1978) conclude that there is little agreement about whatconstitutes a learning disability. In a review of different definitions,Sutaria (1985) suggests that no one definition is satisfiwtory.Kavale and Ft:mess (1985) attribute the problem to the exploratory nature of past and contemporarywork in the field. Although experts disagree about the definititmof learning disabilities. school districts must nevertheless providespecial education to all students who are said to be learning disabled.'11w 1977 Federal Register provides an operational definition: It equates alearning disability with the severe discrepancy between astudent's ability and achievenwnt. But the 1977 federal regulations neither specify tlwsiie of a "severe" discrepancy, nor do they indicate themethod bv which such a discrepancy shouldbe determine(I. The 1977 definition also lists conditions thatexclude students from services in 1.1) programs. Although some of theseconditions can 1), measured quite easily and with great reliabilitv (e.g..visna..mpairmeol!. ,)mers are very difficult to measurereliably (e.g cultural disadvant.t.Ae). Even when the condition itself can be measured,lmwever. educators ,.onwf measure the degree towhich the condition does or does not inAnencethe dimTepancy between a particular student's ability andachievement test scores (Smith, 1983). '11w(lifficulty of making this judgment also contributes to the heterogeneity of thepopulation of studei:ts said to be 1.1).

Discrepancy: brief history and research.The consensus of researt h does not endorse any definition of 1.1), including thatin the federal regulations. Nevertheless. ;nost operational definition.% relv on some measureof the severe discrepancy oet ween astudent's ability and arhievenwnt. This method of identifying I.1) tudents is the most oblectiveavailable mea ;ure of the failure of a normally intelligent child to learn. A severediscrepancy is interpreted as strong evidence that a learning disability marexist. although (:arrier (1983) notes that it is fallacious to assume that all casesof severe discrepancy indicate the presence of neurologkal defects.Nonetheless, since the objective measurement of subclinicalneurological defet ts is not possible, a discrepancy in criterion perform;e is perhaps a reasonable alternative.

2 !I 6 288 Inhibikri Drvelopmput: Underachirvement Bateman (1965) first introduced the notion of discrepancy to definitions of The notion of discrepancy is implicit in the definition of "minimal brain dysfunction" that was developed hy a 1966 task force: 4. children of... average . . intelligence with...disabilities ranging from mild to severe ." (Clements, 1966, p. 9). According to Reynolds (1985), the discrepancy clause was included in Pl. 94-142 regulations because it was the only point tm which amajority of experts could agree. The notion of discrepancy is almost as old as norm-referenced IQ and achievement tests; and there are many' different discrepancy formulas. UnfortunatelY, however, researchers (e.g., Forness, Sinclair. & Guthrie. 1983) have found that different formulas declare different numlwrs and kinds of students eligible. A perhaps more fundamental issue. Imwever, is that some of the formulas aremathenzaticrdhimpropfr. Reynolds (1985) provides the most cogent discussion of* this issue to date. He reports that the only mathematically acceptable formula is the model called tlw "frequency of regression prediction discrepancy" (p. 461). According to Reynolds (1985, p. 461), the model "answers the question of whether there is a severe discrepancy between the achievement level of tlw student in questi(n and all other children with the same IQ.-

Characteristics of gifted children with learning disabilities.Alt hough in theory gifted underachievers shouki not necessarily be considered learning disabled, in practice their depressed performance on measures of achwvement may make them eligible for -ilacement in I.1) programs (cf. Wolf & (;ygi, 1981). According to Setif (1983). most gifted students' achievetnent is significantly below then expected level of performance. Coinpounding the difficulty fdir gifted underachievers is the school's typical insistence that these children master basic skills before they prod ced to more advaiwed milieu' (Whitmore. 1982).Remedial programs designed to improve L.1) students basic skills are not necessarily el fective (Frauenheim. 1978; Ilaskell. Barrett, & Tavhir. 1977), and phu ement in a remedial program may further restrict the giftedchild's access to learning experielwes that are advanced, content-specific. or interesting. Gifted students, whose nwasured intelligence is two or numc standard ! deviations above the mean. are more likely than average students to demonstrate severe discrepancies. For example. a six-and-a-half-year-ohl thilt1 with an IQ of 115 may achieve a 3.3 grade equivalency score on an individual achievement test. This score might correspond to a standard score of 122. The disc repancy between this c bild's IQ and achievement in reading is, therefore, 23 points (approximately 1.5 standard deviations). In many school districts, this degree of discrepancy would be considered severe. Vet. comnum sense suggests that this youngchild's reading achWvement is adequate 11w child's performance is actually' 1.5 to 2.0 rearsabove his or her

2 !4 Inhibited Development: Underachievement 289 grade placement level. The apparent discrepancy may result entirely from lack of preschool instruction in reading or from average perceptual-motor development, or it might in fact be the expected achievement kvel for most children of this age and IQ (cf. Reynolds, 1985). Unfortunately, some teachers might refer such a child to be evaluated for placement in an IA) program. If the child is divergent, disruptive in class, noncompliant, or has poor handwriting, the teacher may make such a referral. According to Bull (1987. p. 6), gifted children with learning disabilities "are usually only referred when they exhibit some signs of social, emotional, or personality disorder." Academic skill deficiency is rarely the reason fOr such referrals. If a teacher chooses to make a referral, subsequent placenwnt in the 11) program is Rely. According to Vsseldyke and Algozzine (1982). special education classification decisions are influenced more by background in- fOrmation provided by the referring teacher than by evaluation data even when evaluation data clearly ccmtradict the background information. These researchers (p. 228) suggest that "examiners hold and seek to confirm (with or without appropriate evidence) preconceived notions about the assessment outcomes based upon the child'srharadniAlics"I'Vsseldyke and Algotzines emphasis]. The characteristics of some gifted children (e.g.. divergence, nonctmh,rmity) may. therefore, increase their likelihood of" lwing identified as learning disabled. Anot lwr, though perhaps less common, situation involves neu- rologicallY impaired (i.e.. organically brain-damaged) gifted students who use conceptual strenwhs to compensate for perceptual deficiencies. 'nurse students may disguise their dd ficulties but mav also not reveal their giftedness (Whitmore. 1982). Hence, the underac hievement of such students may be chronic because they may Ilcrer be provided a stimulating Uttrrk11- lum. his probkm is compounded by the tendency of "teachers .. tobe easily satisfied with wot k of a good standard" (Pirozzo, 1982, p. 19). Be- cause such students do produce work of a good standard and seem to expend considerable effort M doing so. their teat hers may overlook their other chat ac teristics. A teacher is not likely to investigate whether or not such a student's large vocabulary and elaborate oral stories are associated. for example, with exceptionally h;,...;h According to Buil (1987). gifted students who are learning disabled typically exhibit certain characteristics. Frequently they have sloppy handwriting, difficulty in following directions, and poor spelling. The., often make reversals in reading and writing letters and sonwtimes al !". make numeric transpositions. Bull (1987) emphasiies the emotional picthlems that such students are likely to suffer: They are tokl that they are gifted. yet in school subjects they tail to perform. This disparity may produc e a high level of frustration which furdwr impedes sin h students learning. 290 Inhibited Development: Underarhirvement Gifted Children with Emotional Problems Although teachers and administrators continue to concern themselves with the presumed emotional imbalance of gifted children, researchers have tended increasingly to substantiate the normal enunicmal adjustment of the gifted (see e.g., (a lluci, 1988). Gifted children are no more likely than average children to have emotimial problems;nevertheless, a small percentage of gifted children do have serious enmtional or behavioral difficulties. The discussion that fiillows briefly considers the nature of emotional disturbance. particularly when it is viewed as deviance. A review of the characteristics of those gifted children who are emotionallY disturbed also f011ows.

Deviance as a measure of emotional imbalance.Since gifted children are identified because of" their deviant scholastic performance or potential. schools may be particularlY anxious to ensure that these students do not behave in sociAlv or emotionally deviant ways, 'this concern is demonstrated in spite of the general trend of rcient research, which substantiates that the gifted are not at iish emotionally. Children who have serious enuitional problems have a difficult time adjusting to social expectations. Thev may have a great deal of difficulty getting along with others. Their self-concepts may be poor. and their interpersonal relationships strained. Students who truly have deviant pattet n:. of behavior (I) display behaviors that are inappropriate for their chic mo- logical and mental age. (2) exhibit such difficulties over an extended period of time, and (3) fail to respond to parents' in- teat hers attempts to modify their behavior. According to Ndsoa (1985). such c hildren display disturbed behavior. not .just disturbing bdiavior. UnfOrtunately, some teachers are unable to distinguish between disturbed and disturbing patterns of behavior. One study tumid that class- rocnn teachers perceived 20,4 pen em of their students as exhibiting behavior disorders (Kelly, Suffix k, Sr: Dykes, 1977). In this study, teaulwrs also identified boys and blacks more frequently than girls and whites (twice as often. in each case). Apparently, these teat-hers had a verv low tolerance for deviance. The l'efiavior of gif led t Illicit en ts Amin itial in some was s. Their responses to questions may be divergent, and they may t hallenge autlunity more of ten than average children do (Torrance, I 9(i3).When gifted t hildren demonstrate these atypical patterns of behavior. they may at risk for classification as emotimally disturbed.

Characteristics of gifted children with emotional problems.Most studies link low IQ with emotional disturbance and dermineituv (e.g.. White.

29:4 InhibitedDevelopwwitt: Underachievemnd 291

1966); nonetheless some studies haveinvestigated the relationship of social and emotional disorders with highIQ (e.g., King, 1981; Morse, Cutler. Sc Fink, 1964). In fact, many studies(of both types) do include subjects with IQs in the superior range (Kauffman,1981). There seem to be no systematk studies of thecharacteristics of gifted children who are also emo6onally disturbed.Such studies would, in fact, be very difficult to conduct,because locating a sample of disturbed gifted children would be a great challenge. IdentificationOf emotional disturbance entails quite a hit more than theadministration and interpretation of a test or battery of tests.Often, the diagnosis is based on the clinical expertise of a psychologist or psychiatrist,and the judgment of one clinician may diner considerably from that of another.Even itchnicians were to agree On idl of the diagnoses, differem childrenwould still manifest their disturbance in a variety of ways. vlence, any groupof children who were classified as emotionally disturbed vouldprobably differ I rom any other such group, and individuals within any otw groupwould also dif fer greativ from One anollwr. Compounding the problem is the difficuhv inidentifying the giftedness of seriouslydisturbed children. in many cases, these childrenstalents are hidden. Evensituationa; anxiety can depress children's mores on tests of intelligence and achievement. The seriousemotional disturbances of sonw children winobviously interfere with their ability to demonstrate talents. I InfOrtunately, if' severe emotionaldisturbatre lasts for many years during ilw devehymental period, the child'scognitive devehynwin ma% Iw seriously impeded. Such children maysuffer permanent loss of abilities that they might otherwise have developed munefully. Although there is no research about giftedchildren who are emotionally disturbed, there is much speculationalsmt the topic. Many educators and psYchologists express concernalxnn the stress that gif led students are likely to ent ounter.Foster (1985. pp. 142-144) presents thefollowing list of adjustment problems that gifiedstudents often must deal with:

inherent tension between the unusual expectation%and tweds of Ihe gilled child anti the normative standaids of the larger sotjet%. 2.the child's awareness of issues prim to havingthe resotutes with whit h to resolve them. nd 3,intdlet !nal iudependeme ficnn patents andnrtmethate tannic that !net edes emotiomal and social independence.

According to Foster (1985), these diffuuhies can be handkd through school counseling and guidance programsfor gifted students. But this recommendation suggests that Foster is discussingonly the minor adjustment problems that giftedstudents sometimes encounter. 'Vilest. problems 292 Inhibited Developmeta: Undeschievemient are transitory ard (Alen disar .ar as children mature. Khatena (1978), on the other hand, oxpresses concern that the situational stress that some gifted children ext.wrience will lead to serious disturbance:

Mro longed enforced repression of the creative chikl's !leech ir rad to emotional pt. :^olems and neurosis, and even psychosis. Neurosis...is a condition generated by acute and prolonged anxiety states....Neurosis hinders rather than facilitates the functioning of the creative process... .The creative individual who tws his pr:xiuctivity blocked may develop behavior traits similai to those of psychotics...his behavior might become withdrawn or schimphrenic. (lthatena, 1978, pp. 89-90)

Khatena'a conclusions should be viewed cautiously, however, since they are based on speculation rather than on empirical study. Considering the evidence that shows the relative emotional health of gifted students, it seems reasonable to assume that gifted children will weather situational stress at least as well as average children. l'he more serious emotional problems that some gifted children like some children of average abilitywill face appear to be a different matter. "l'hese types of disturbance are likely to have the following mattifes- Cuins:

1. They may involve inadequate ego development. 2.'1 hey may invoke birarre or self-destructive behaviors, 1. they may relict t serious probkms of family interaction. 4.Thev may be resistant to change. They may in.erlete with the child's intel action with rithet mental-age peers. ti. I hey- may limit the IOM's ability to interact with minks.

These are the sorts of problems that may require therapy with a psychiatrist. clinical social worker, or psychoh)gist. Some critics question the value of therapy in treating ;wine or chronic emotional problems. Such critics cite institutional recidivism, the difficulty of working with schizophrenic patients, and organic deficiencies as evidence that explains why therapy is ineffective. Nonetheless. .,ifted StIldents, who so typically possess ackat n et! y erbal skills, mas actually be more likely than other individuals to benefit from forms of therapy based on "insight.- Such students may be in a good position to articulate their conflicts, to analyze their past patterns of be- havim,and to plan better courses of action fOr the future. Readers should note that acute psyclmlogical problems (i.e.. those associated with a crisis in an otherwise functional life) are more amenable to "insightful" therapeutic intervention than chronic. problems (i.e., those associated with an ongoing problem that hinders a person's ability to function). Very bright students 3 1 Inhibited Development: Undernehirvement 293 with chronic, disabling emotional problems maybenefit front a combination of insight therapy and the kind of behaviormodification recommended for many other studentsoften with average1(2s or below-average 1(2swho have behaviordisorders. In the case of both acute and chronic problems, it is critical that students be placedin the hands of expert thei apists. Poor therapy can contribute sigMficantly tostudents problems.

SUMMARY

This chapter presented theories and researchalxmt the underachievement of gifted children. By examining varioushypotheses that attempt to explain underachievement, is provided a frameworkfot evaluating the different empirical studies. Two kinds ofunderachievement were considered: ( I) underachievement represented bylow grades and (2) underachievement represented by low scores onstandardized achievement tests. This latter type was associated with certain handicappingconditions that impede the performance of bright students. In reviewing the literature on the characteristicsof gifted students who get poor grades, the chapter consideredschool behaviors, personality traits, and family dynamics that might contribute tosuch students' underachievement. It reported the inconclusive, and ofienconflicting, evidence provided by such research. The discussion of gifted students with handicapssuggested that stereotypes of handicapped childrenmight limit parents' and teachers' expectations for such children's performance.Cautions Nere given about the effects of lowered expectations on such students'achievement. This chapter concentrated on the underachievementof individuals - underachievement that often can be explained in termsof psychoanalytic, psychological, or biophysical etiologies. In mostinstances, the underachievement of such individuals can be attributed either totheir handitaps, to peoples' responses to theirhandicaps, or to complex interactions between home, school, and personalityfactors. The more systematic underachievement of group al individualsunderachievementthat nmre strongly plflects social and economic influencesisconsidered in the next chapter. .11 Unfair Discrimination, Poverty, and Minority Gifted Children

I. Focusing Questions II. Fact: Discrimination Exists Ill. Effects of Discrimination A.Social Class B. Social Mobility and the Culture of Schooling 1. Expectations 2. Internalizing status expectations 3. Discipline problems and the life of the mind IV. Minority Gifted Children A. Group Differences and Intelligence 1. Verbal ability and minorities 2. Blacks and scholastic achievement 3. Ability patterns across ethnic groups V. The Relationship of Gender to Patterns of Underachievement A. Risk Factors for Boys 1. Biological factors 2. Environmental factors B. Gifted Girls and Women 1. Scholastic achievement 2. Low math achievement 3. Achievement and compliance 4. Adult achievement of gifted women Vt.Gifted Students in Rural Schools A. Rural, Small, and Isolated Schools: A Definition B. Characteristics of Rural Gifted Students 1, Identification problems

294 3o3 Unfair Discrimination, Poverty, ond Mbwrity GiftedChildren 295

B. Characteristics of Rural Families and Communities 1, Characteristics of rural families 2.Characteristics of rural communities 3. Characteristics of rural schools VII. Summary

Focusing Questions 1. How are students from different social classestreated by their classroom teachers? In what ways does this treatment relate to the future work roles that the students are expected toassume? 2. Why are so few minority-group childrenidentified as gifted? Is it possibleusing standardized teststoidentify more minority gifted children? 3.What differences in children's cognitive styles canbe attributed to their ethnic backgrounds? 4. Why are boys at greater risk than girls ofbeing categorized as learning disabled? 5.How does 1:ierformance in mathematics reflectdifferences in the ways that boys and girls are socialized in theirhomes and at school? b.What charaderistics of rural families, communities,and schools contribute to the underachievement of bright childrenfrom rural areas?

FACT: DISCRIMINATION EXISTS

Unfiiir discrimination in educational practiceexists. it existsdespite the fact that most citizens look to the schools toprovide a way fOr students to improve their social andeconomicstanding. ln fact, unfair discrimination exists in schools in spite of many educatorseffo _s to disable its effects. We use the term "unfair discrimination"throughout this chapter to highlight pedagogical concern for justice. Somediscrimination is fair, as when gifted students are discriminatedfrom other students for the purpose of providing them with achallenging program. in general, "discrimination" means "making distinctions." Somedistinctions are contrary to justice. This chapter investigates such distinctions asthey affect gifted students. We refer to such distinctions as "unfairdiscrimination". Evidence ofunfair discrimination in educational practiceabounds. Given such well-documented facts, it is no wonderthat students in poor neighborhoods perform so badly. Three categories ofevidence about discrimination in schools are indicated as follows:

Fint, srlwas are fgregated by race and clas.s. Schoolslocated in regions or neighborhoods where income or wealth is low 296 Unfair Discrimina him, Poverty, and Minority Gifted Children

(e.g., in ethnk or racial ghettos) receive much weaker support than schools in affluent regions or neighborhoods( Jencks etal.,1979;Mosteller & Moynihan, 1972). Serond, some schools do not teach thinking and reasoning. Students in segregated schools that serve the working class and underclasses experience a kind of learning different from their peers who attend more affluent schools. In working-class schools, students experience a regimen of rote learning and training in submission to arbitrary authority (Anyon, 1980; Bowles & (3ititis, 1976; Wilcox & Moriarity, 1977). Third, most of these .schools are not effertive. In these schools, fbr example, the expectation For student achievement is low, use of instructional time is poor, and salool climate Ls hostile (oftendomi- nated by violence and vandalism). Finally, students in these schools learn less than students who attend more affluent schools (Coleman et al.. 1966; Oakes, 1985; Sizer, 1984).

Are educators to blame for such practices and such results? Only in part, for the source of such practices and resultslies in how society is structured in general. Nonetheless, when the educational systemand when individual educatorsknow what needs to be done to promote learning, but fail to take the necessary steps, unfair discrimination is enabled. Schools necessarily reflect the predominant values and social structures that characterize the societies of which they are a part.Schools, after all, exist to perpetuate the cultural heritage. They may, however, also thape the cultural heritage as they transmit it to students. Some of this shaping occurs by chance, some by choice. For example, some schools in impoverished neighborhoods are 'effective." Considering their social and economic circumstances, these schools produce higher levels of student achievement than would be expected. In addition, some teachers in less effective schools do succeed in teaching their students to learn and to value learning. Such schools and such teachers are rare, however, because fighting unfair discrimination requires positive action. It does not happen spontaneously in our culture, which endorses many kinds of unfair discrimination. Fortunately, we do know a good deal about the targets of unfair discrimination in school, and we know a little about the measures that can be used to limit the effects of discrimination. For the most part, the targets of discrimination in gifted education are not surprising. Gifted black, His- panic, and working-class students are most at risk. But various cultural stereotypes place many girls and some boys (of all classand ethnic groups) at different sorts of risk. Geography can also beused to identify the targets of discrimination; the distinction 1 yen town and country (urban and rural) is, for example, notable. Since t:ounts (19:32) wrote on the topic, this distinction has been largely ignored in gifted education. Itis, however, receiving more attention today. thsfair Discrimination, Poverty, and Minority Gifted Children 297 Filially, educators should remember that the effects of discrimination are cumulative. Unfair discrimination is carried out insistently, and the -ffects of discrimination accrue steadily. In practical terms, this observation means that the effects of discrimination are themselves sources of further discrimination. For example, impoverished students start school with achievement levels not so very different from their more advantaged peers. But as they progress through school, they fall further and further behind. Those who fail to become marginally literate will, as adults, suffer additional discrimination on that account.

EFFECTS OF DISCRIMINATION

Unfifir discriminafion contributes to the division of members of society into different socioeconomic classes. The influence of this division is so significant that socioeconomic class is one of the most common control variables used in educational research. It is identified nearly as frequently as age and gender as an important source of variability in students' academic achievement. Efforts to understand and control the effects of 3ixial class are viti- ated, however, by limitations in the terminology by which it is described. The terminolog7 used to make comparisons between socicoeconinnic classes is often vague and misleading.

Social Class Williams (1976) describes the history of the word "cla.ss" as a term for social groups. He notes that the term became widespread during the In- dustrial Revolution. During tbis time, "middle class" was coined to describe a position between the aristocracy and laborers. At about the same time, thc terms, "productive classes" and "privileged classes," were used to distinguish between people who earned a living and the aristocracy. By the 1830s the term "working class" was used to distinguish between lalxirers and other economic groups. These uses of the word "class- are based on two different perspectives. The upper-, middle-, and lower-class division of society is hierarchical and assumes a continuum. Upper-class families have more money and more prestige than middle- and lower-class familks. But. the term "working class" distinguishes on the basis of production and does not imply an ordinal scale. Most research is based on arbitrary assignment of students to upper-, middle-, or lower-class status. Lack of precision in these terms means that research that claims to control for social class does not always do so. The term "middle class,- for example, covers a broad range of lifestyles. In an

3 6 298 Unfair Discrimination, Poverty, and Minority Gifted Children analysis of the income of the U.S. population, Rose (1986,p. 9) profiles two "middle-class" families:

In one the husband works as a forklift operator at an assembly plant while the wife stays at home with the two children; total family Mcome equals $18,000. In the other, a st:hurban couple with a $150,000 house, the husband isa demist, the wife a psychologist; total family income is $90,000, andone child is in college and the other attends a local private school. Both of these families might describe themselves as being "strapped" for money andas having little left over for frills. Both would probably consider themselvespan of the "middle class," the first family perhaps adding the adjective lower" and the second adding "upper."

Rose contends that the category "middle class" isso broad that its use obscures important political and economic distinctions. The differentre- sources of families such as those described previously lead them to approach educational issues differently as well. The prospects of the children from the first family are dependenton the quality of the local schools, but those of the children from the second familyare not. Even when they attend public schools, the children of higher income parents haveaccess to private lessons, artistic performances, iravel, and other experiences that are unavailable to many "middle-class" children. For educational research and practice, social structure is perhaps better defined in terms of discrete structural segments that relate to theecono- my. Social class interpolated from rank on a single dimension is flawed for several reasons: (1) classes must be defined by relative (and arbitrary)distinctions between low, middle, and high ranks; and (2)a continuous socioeconomic scal,- does not account for the conflicting interests of different classes (Poulantza1974; Wright. 1979). In discussit% , he need for more precise terminology, Labov (1972) had the following comments:

.:he term "lower class" is frequently used as opposedto "middle class" . . it is useful to distinguish a "lower class" group from working class. Lower- dass families are typically female-based...with no father present to provide steady economic support, whereas for the working class there is typicallyan intact nuclear family with the father holding a semi-skilledor unskilled job." 181)

Although it should be recognized thatmany working-class families are female-based, the distinction that concerns Labov is an importantone. There are major differences in the kinds of environments of children usually categorized simply as "lower class." Recently, theterm "underclass" has been used to describe that portion of the population that hasno wealth, no income, and no prospects of either (e.g., Lehman, 1986; Mann, 1986). Twenty-two percent of children in the U.S. live in households with incomes 3 7 Unfair Discrimination, Poverty, and Minority Gyted Children 299 below the poverty level of 511,000 established by the Social Security Ad- ministration (Rose, 1986). Descriptions of the living conditions of children who belong to this class have been givensome attention in the popular media, but, since the 1960s, severely economically deprived children have been virtually ignored in the education literature. Nonetheless, the number of children who belong to this class has been growing ata rate that alarms even the representatives of business and industry (e.g.. Committee for Economic Development, 1985). The bright children of the underclass have almostno chance of being identified as gifted except through the use of quotasor special methods of assessment, such as Mercer's (1981) System of Multicultural and Pluralistic Assessment (SOMPA), that take into account debilitating social andeco- nomic circumstances. No study has been conducted to determine how many children from this group have been identified as gifted. However, using Terman's data, we can calculate the chances thata child from this group has of being so identified: I chance out of 150. Of the 1500 gifted subjects, fewer than 10 came from families of laborers (Terman. 1925).

Social Mobility and the Culture of Schooling Given an average IQ, the child from a middle-class family is 25 times more likely to end with a well-paying, high status job than is the child from a working-class family (Bowles 8c Gintis, 1976). Among the factors that make it so difficult for working-class childrento improve their socioeconomic status are (I) a limited education and (2) the inhibiting attitudes that result from their economic circumstances. These attitudesare originally imposed externally (e.g., at school and home), butmost older children and adults have internalized them. Advocacy of equal educational opportunity is basedon the belief that education enhances social mobility. It is true thatmore schooling offers minorities enhance0 status and a closer approximation of economic equality. The economic return fbr schooling, however, is much less for children from low-income families and for blacks than itis fbr children from middle- and high-income families (Bowles 8c Gintis, 1976; Wright, 1979).

Expectations.Even under conditions of apparently equal educational opportunity, social forces undermine the academic efforts ofpoor children. As Good and Bre phy (1987) report, children from lowerso- cioeconomic status families are not always treated equitablyin the classroom. Important, too, is recognition that membership ina lower socioeconomic group, whether underclass or working class, works against cultivation of the attitudes that are associated withsuccess in school. Ac- cording to Kohn (1977, p. 189), "The essence of higher class position is the expectation that one's decisions and act;onscan be consequential; the es- 300 Unfair Diwrimination, Poverty, and Minority Gifted Children senceof lower class position is the belief that one is at the mercy of forces and people beyond one's control, often, beyond one's understanding." Parents' expectations for their children are in fact influenced by their class background. Working-class parents report that they value obedience, manners, and cleanliness. They want theirchildren to be "good" students. Middle-class parents, on the other hand, report that they value qualities such as critical thinking and curiosity. The former qualities represent obedience and external motivation; the latter imply independence and self- direction. It is no accident that the high expectations of teachers appear as a significant influence on the achievement of children. It is also no accident that. to pronune high expectations in working-class or underclass schools is very difficult. There is much in the structureof' society that undermines high expectations for such children: Cultural, economic, and familial practices can determine what typically happens in schooL. Schools that do promote high expectations for these children, however,provide the essential counterexamples that demonstrate that it ik possible for such children to be successful learners. As Edmonds (1979) notes, we can assurethat these children learn well; the question is, do we want it to happen?

Internalizing status expectations.'1'hecoliditions of txT upafional life at higher social class levels also) foster thebelief that the attainment of individual goals is possible. Professionals and managers learn to expect intrinsic as well as extrinsic rewards from their jobs. The conditions of occupational life at lower social class levels, on the other hand, limit the worke.view of a job to the extrinsic benefits it provides. Unskilled and semis! 'lied jobs foster a narrowly circumscribed conception of self' and society and promote the positive valuation of conformity to authority (Kohn, 1977). The same conception of self and society is taught. via the "hidden curriculum," in working-class and underclass schools (Anyon, 1980). The type of education that working-class andunderclass students receive also seems to contribute to the development ofattitudes that inhibit their self- direction. Schooling prepares them for roles as compliant participants in the labor force. The differences between instruction in the schools attended by working class children and those attended by the children of parents who earn a high income have been documented in many studies (e.g.. Wilcox, 1982). Upper class children are schooled fir self-direction, and lower-class children are treated in ways that cause them to rely on external reinforcement for learning. Identification with the middle class is an integral element of the teacher's role. By their expectations, attitudes, and behaviors, many teachers, unknowingly, contribute to the rigidity of the class structure. One study, [Wear Discrimination, Poverty, and Minority Gifted Children 301 for example, found that teachers' placement of studentsinto reading groups was more highly correlatedwith the children's social class membership than it would have been if they had been placed strictly onthe basis of achievement test scores (Haller & Davis, 1980). Placement in a low orhigh reading group is not only a respcmse to social class differences,it also perpetuates them.

Discipline problems and the lye of the mind.Children who sufkr unfair discrimination may seem to be hopelessly caught in asocial structure and a culture that is bound to cripple their access to thelife of the mind. Giroux (1983) pMnts out, however, that rebellion and resistance toauthori- ty exist as latent reactions tosubmission to authority in these circum- stancesamong both adult workers and students.The discipline problems so connnon in working-classand underclass schools attest to this fact. ('iroux asserts that the resistance that these students practice is a token of their submission. He advises teachers to find ways toshow students how the form of their resistance (e.g., acting out, disciplinepmbk.nts) puts them further under the swayof discrimination and oppression, and how it actually makes oppressed students more likely targetsfor exploita- tion. Giroux believes that through opening up this topicfor ongoing dialogue among teachers and students, students can begin tounderstand where society has placed them and what their responses might be. Giroux (1983) dearly believes that it is the role of teachers to reveal tostudents in these circumstances the liberating power of knowledge, language,and re- flectlcft.

MINORITY GIFTED CHILDREN

Front a biological viewpoint, there is every reason to expect the santeinci- dence of high ability in children from racial and ethnic minorities asin racial and ethnic majorities (Gould, 1981; Lewontin et al., 1984);but the chances of' minority group children being identified as gifted are extremely low. With the exception of children from Jewish or Oriental families,children from minority groups are grossly underrepresented in programsfor the gifted. The fact that so few minority students are identified asgifted is usually attributed to ( 1 ) biases in the content and pnxedures of IQand achievement tests and (2) cognitive difTerences between the minorities and the white, middle-class majority. Despite the belief that minority gifted children can and should be identified, methods of identifYing gifted students have changed very little in the past 60 years. The problem of' identifying bright minoritystudents persists. Teacher referral remains the first step in entry to thegifted program. An IQ cutoff in the 97thpercentile or higher still serves as the

3 I I) 302 Unfair Discrimination, Poverty. and Minority Cy 'led Children primary criterion for placement in a gifted program, just as it did in the first large-scale effort (Terman, 1925) to identify gifted studen.s in the United States. Many of the provisions intended to overcome the effects of discrimi- natory tests and identification procedures make use ofsubjective measures. The subjective nature of such measures (e.g., teacher checklists) means that their predictive validity is questionable. The most promising methods for identifying gifted minority students, that is, those methods that take into consideration students' lack of educational opportunity and yet incorporate measures that are good predictors of academic success, are(1) quota systems and (2) the establishment of local norms.

Group Differences and Intelligence Using socioeconomic status as a proxy for social class, studies have fbund that social Class and intelligence correlate moderately, about a 0.50 Pearson product-moment correlation coefficient (Howley, 1982). Some researchers (e.g.. Gardner, 1961) have argued that this degree of correlation is evidence of the efficiency of our social system in sorting individuals on the basis of aptitude. Other authors argue that this degree of correlation demonstrates the effects of a social system that attempts to strengthen the relationship between social class and cognitive performance (e.g.. Lewontin et al., 1984).

Verbal ability and minorities.'The close association between superior verbal ability and high IQ test scores has implications for the identification of bright minority children. According to Davidson and Balducci (1956), gifted children from families of high income usually have a more advanced vocabulary than gifted children from families of bw income. These researchers found this diflerence, even though the low income children had a slightly highrr average IQ. In order to score in the highest percentiles on most IQ tests, a child usually needs to have a large vocabulary. Working-class and underclass children are at a disadvantage; their vocabularies may not be as large as those of middle- and upper-class children. Sometimes their vocabularies though largeinclude many words and expressions that are not standard English. Differences in the language used at 'some and in the community may account for limitations in the standardEnglish vocabulary of some minority students. Others may fail to absorb or use the vocabulary of' the classroom because they perceive the hostility of that environment. In an ethnographic study of classroom behavior. Heath (1982) observed the response of a kindergarten-age black child to activities in the classroom. Although there are a number of possible explanations for this child's behavior, the behavior itself may adversely influence the child's vocabulary development:

3 1 Unfair Discrimination, Poverty, and Minority GOid Children 303

During those school activities which focused on giving labels to things and naming items and discussing their attributes, Lem did not participate. He listened and often tried to escape these structured sessions to play with trucks, puules, and so forth. He had no interest in looking at books and being asked questions about them.. .. (p.121).

Heath's explanation of this child's behavior is based on differences between the linguistic style of working-class black families and the linguistic style of the middle-class white teacher. The differences, she believes, makes the black children uncomfortable and uncertain about the proper responses to questions their teachers ask. Labov (1972) addresses an issue that Heath's explanation ignores: Black children confronting white teachers or psychologists are, on some level, confronting an enemy. Labov acknowledges that most observations of the language behavior of black children have found relatively barren language. However, he sees this numasyllabic behavior AS defensive; it is produced by placing the child in a situation in which "anything he says can literally be held against him" (p. 185). Under the circumstances, the minority children's efforts to second- guess the questioners and their reluctance to speakspontaneously are understandable. This view offers a possible reason why, as the teachers in Heath's study report, the children just stare at them even when it seems that they could answer the question. The teachers' impression is consonant with Latxw's opinion that minority children's defensive responsesto threatening situations sometimes causes them to seem "inept" (p. 191). Because of the circumstances under which observations are made, however, many samples of black and working-class children's language are of questionable validity. The inferences based on samples of such language are even more problematic. Bernstein (1961), for example, attributes poor academic performance of economically deprived school children to their deviant language, which because of co tain traits, results in deficits in their capacity for complex reasoning. He describes the language of uneducated perm ns, regardless of' race, as inadequate for discussing abstract ideas. Amonghis generalizations about the language of working-class and underclass individuals are the following:

Language Characteristic Resultant "Deficit" repeated use of a small number of logical qualifications cannot be ccmjunctions (and, because, so. then) expressed exactly heavy reliance on implicit meaning cognitive and affective differentiation reduced frequent use of implied questions further analysis and search tor (i.e., tag questions) at the end of reasons discouraged statemel, (e.g.. "isn't it?") 312 304 Unfair Diserinsiviotion, Poverty, and Minority COrti Children

The inferenct s that Bernstein draws from these language characteristics are untenable. Logic systems usually relyon only a few conjunctions. In addition, many authors would disagree with the assertion that heavyre- liance on implicit meaning reduces cognitive and affective differemiation. Also, one of' the characteristics typical of middle-class mothers' speechto their children is the use of many tag questions. This characteristic doesnot seem to damage middle-class children's analytic thinking. Labov (1972) counters claims such as Bernstein's by demonstrating that nonstandard English is as logical as standard English. Inccotrast to the deficit theory of' hneer-dass language, Labov pointsout that much middle-class language is dysfunctional. Richard Mitchell (1979) in Less Than Words Can Say provides many telling examples tosupport this claim.

Blacks and scholastic achievement.Economic deprivation is the major Atte to consider in efforts to identify minority children with superior academic ability. Measures of intelligence correlatemore with economic status than with skin color. The majority of economically deprived children are, of course, white (Martinson, 1972). Nonetheless, among blacks, a far greater proporthm of children are poor. In 1982 less than 66 percent of' blacks were employed (Shiels, Weathers, Howard, & Givens, 1982). Black. female-headed families represent a growing proportion of poor people in America: from 7.3 percent in 1959 to 19.3 percent in 1984 (Lehman, 1986, p. 33). While the median income fortwo-parent households is $28,300, the median income for households headed bya woman (i.e., either white or black) is $13,500; and almost half of the black children in the United States live in single-parent households headed bya woman. It is common knowledge that black children, asa group, have a relatively low mean score on IQ tests. In the United States, however, it is less well known that the mean IQ score of poor children, regardless ofrace is below the mean (Gould, 1981). When middle-class black studentsare compared to the norm on IQ tests, their mean IQ isaverage (Jenkins, 1950). The disproportionately large number of poor black children is sufficient to explain the lower mean IQ score of black children. As mentioned earlier, however, controlling for social class is not alwaysan effective means of insuring the comparability of students' backgrounds, and the results of studies with "matched" samples are mixed. According to Jenkins (1950), educators' preoccupi,tion with black children's low mean IQ score has diverted attention from the within-group variability of IQ. He points out that among middle-class black children, the incidence of recognized giftedness is as highas among middle-class white children. Referring to his study of highly gifted black students from several major cities, he also points out that theupper limits of intelligence tests are not the exclusive domain of white children. The black students in his study had IQ scores ranging from 160 to 200 ( Jenkins, 1943).

3 :3 Unfair Diserintination, Poverty, and Minority Gifted Children 305 Jenkins located no highly gifted black children from rural areas. Studies from the 1940s found that rural blacks from the South had even lower IQ scores than blacks raised in the North ( Jensen, 1973). This effect seems to persist among recent black migrantsfrom the rural South. On the basis of a study of blacks in Chicago, Lehman (1986) reports that the blacks who make up the underclass there are those whose parents immigrated from the poorest and the most rural areas in the South. In these rural southern areas, throughout the 1950s a system of segregation and share- cropping prevailed. Most blacks were denied the right to vote by a poll tax and a literacy test. Even a junior high-school education was rare among rural blacks. Black rural school systems in the South still face much more selious problems than face other rural districts (Nachtigal, 1982). Simply moving out of the South to a northern urban area was enough to improve some black children's IQ scores(Lee, 1951). Lee's study of southern black children who moved to Philadelphia found that the IQ scores of the southern-born childrenwho entered the school system improved as they progressed through the urban system. hose who entered the schtml system in the first grade showed a greater gain, apparently because they had been in the urban schools longer; but the rate of gain matched that of children who moved to Philadelphia later in their school careers. After several years in the Philadelphiaschool district, these children's IQ scores were equal to the IQ scores of Philadelphia-liorn black children. Black children who score in the gifted range on IQ tests often ccrnt: ffom middle-class families and attend better equipped schools than those available to most black children (Martinson,1972). However, black achievers also come from families whose occupational level could be categorized ab upper lower-class (Schultz, 1958). Their parents tend to be clerical, skilled, or semiskilled workers with relatively low incomes. Regardless of social class, the parents of high-achieving black students apparently support their children's achievement in a number of ways. Ac- cording to Shade (1978), the parents establish close family ties, use connol mechanisms that include moderate amounts of praise and blame, give assistance in schoolwork, establish goals for performance, expressinterest in their children and encourage them. Black achievers score high on traditional measures of intelligence, especially in verbal ability and analytical ability. The personality characteristics of black achievers suggest that they have positive views ofauthority figures, that they are self-confident and goal-oriented. They seem tobe willing to conform to adult demands, and they are cautiousand controlled (Shade, 1981).

Ability patterns across ethnic groups.Some researchers have found differences in the ability patterns characteristic of various ethnic- groups.

314 306 Unfair Diserionination, Poverty, and Mum*" GOrd Children Lesser, Fifer, and Clark (1965), for example, found ethnicdifferences in both overall level of ability and pattern of performance onvarious tests. The Jewish and black groups scored highest in verbal ability;Chinese Americans and Puerto Rican children scored highest in spatial ability.So- cial class had marked effects on the level of performance butdid not seem to influence the pattern of abilities. A study by Backman (1972) found Jewish subjects to berelatively superior in verbal knowledge and mathematics; non-Jewish whitesubjects displayed a relatively flat profile; the black subjects scored highest onitems measuring perceptual speed, accuracy, and memory, and theOriental- American subjects scored highest in mathematics. However, in tnisstudy, sex was a better predictor ofability scores than ethnicity or st ti class. The relative superiority of Jewish children's verbal and numericalability in comparison to their spatial ability was also found in aCanadian study (Marjoribanks. 1972). Some researchers attribute the different ability patterns todifferences in the cognitive styles ofindividuals from ethnic minorities (e.g., Buenning & TollefSon, 1987; De Leon, 1983). In discussing theabilities of Mexican-American children, these researchers suggested that culturally determined cognitive styles distinguished between minority and mainstream children. They found that giftedchildren from Mexican-American homes often used a field-dependent method of problem solving,whereas mainstream children often used a field-independent method. These researchers asserted that field-independent responses would result inhigher scores on IQ tests. Hence, even thebrightest Mexican-American children would score lower on IQ tests than their mainstream counterparts.Because they learn to give answers that take account of the social context,Mexican- American children seem to be penalized on tests requiringindependent, abstract reasoning (De Leon, 1983). According to Tonemah (1987), gifted children from AmericanIndi- an homes also differ frommainstream gifted children, lie believes that American Indian children are often excluded from gilled programsfor the f011owing reasons:

I.assessment procedures tail to account for tribal culturalinfluences; 2.standardized tests are biased in favor of children from the larger society; 3.mime Indian students are unprepared torespond to the types of items that appear On standardized tests; 4. some Indian students have a negative attitudetoward test-taking; 5.indian students are not included in the norming sample for most standardized tests; and 6.standardized tests do not accuratelY reflect the entire spectrum of intelligence or achievement. (Tmemah, 1987, p.183)

3 1 5 Unfair Discrinnsiation, Poverty, and Minority Gifted Children 307 Although research on the topic is not extensive, it does seem that ethnicity can influence patterns of cognitive perfOrmance to some unspec- ified degree. The types of behavior that are most valued by a particular subculture seem to be reflected in minority children's patterns of perfOr- mance. Subcuhures that value artistic performance, for example, cultivate the artistic abilities of children (Tonernah, 1987). Those that emphasize group rather than individual performance condition behaviors that may preclude hign-level performance on tests of individual intelligence (Buen- fling & Tollefson, 1987). Even though cultural differences undoubtedly influence IQ and achievement test performance to some degree, they may have a much less powerful influence than other factors, such as income, wealth. status, and expectations. Historically, ethnic minorities have suffered discrimination that has limited their attainment of income, wealth, and status. Hence, the effects of ethnicity are difficult to distinguish from the effects of poverty or of limited educational opportunity.

THE RELATIONSHIP OF GENDER TO PATTERNS OF UNDERACHIEVEMENT

Oddly enough, both boys and girls are subject to a variety of influences that make them differentially at risk for underachievement. In the case of boys, the influences that lead to underachievement might be viewed as "interaction effects." That is, in the case of at-risk males, a combination of gender* andother characteristics produces an interaction that is associated with underachievenwnt. Perhaps the most notable of these interaction effects is the interaction between gender and race; and as noted above, black males are at greater risk in schools even than black females. Among the population of gifted students, however, girls are presumed to be more educationally disadvantaged than boys. This presump- tion is based on the understanding that performance expectations for girls are not universally high. flomemaking is, for example, still an acceptable vocation for a young woman. At the same time, homemaking is not a valued work. Both child care and housecleaning are tasks that are poorly paid when not performed for free by a homemaker. Men who undertake homemaking are apt to be regarned as very peculiar, if not dysfunctional. Despite the influenceof sexism, attitudes toward the accomplishments of womenandtheir right to be as assertive, ambitious, and self- nmtivating as men, are steadily changing. Educators who work with bright students have a role to play in helping students achieve better by overcoming the constraMts of sexism and sex-role stereotyping.

3 I 308 Unfair Distrisnination, Poverty, and Minority Gyied Children Risk Factors for Boys A number of studies have found that more boys than girls areidentified as underachievers (Pringle, 1970). In her study, Pringle (1970) found that the ratio of underachieving boys to girls was 3 to 1. Among those underachievers who are identified as learning disabled, the overrepresen- tation of boys may be even more striking. Harness, Epstein, and Gordon (1984) estimate that the ratio of LD boys to LD girls is 4 to 1. Phipps (1982) reports that in some LD samples 85 percent of thestudents are boys. Although there are several hypotheses that attempt to explain why more boys than girls are identified asunderachievers, there is no consensus. Some researchers in the 1940s, I950s, and 1960s speculated that biological differences betwee boys and girls accounted for the phenomenon; others blamed cultural expectations and environmental conditioning (Prin- gle, 1970). More recently, educators have speculated about the overrepre- sentation of boys in LD programs. In her review of the research, Smith (1983) cites hypotheses based on medical Mctors. maturational factors, sociological factors, and brain organizational factors. According to Miles (1986), three variables might explain the phenomenon: (1) teacher bias, (2) achievement differences, and (3) gender differences in brain function. Not much research has been conducted to support any ofthese hypotheses, however. Findings from many of the correlational studies tend to be contradictory, and current research on brain functioning is of questionable value. Research techniques for studying hemispheric differences in brain functioning are primitive, at best. At worst, they are based on unsound theories that are derived from the study of small samples of neu- rologically impaired subjects (Springer & Deutsuh, 1981). The fact that LD students are identified in many different ways also makes study of this question extremely difficuft.

Biological factors.Some researchers relate biological differences between boys and girls to the gender disparity in LDidentification. Several studies suggest that 1,oys are at greater risk for birth trauma than girls because newborn boys have higher birth weights and larger heads than girls (e.g., Silver, 1971). The birth difficulties that occur in the labor and delivery of some boys may result in their subsequent learning problems. Other studies report that boys lag behind girls in all areas of matura- tion (Ames, 1968). According to these studies, boys are often unrady for the school tasks that confront them in early childhood. Their frustration and poor performance in the early school years ccmdition later patterns of underachievement. According to some research (e.g., Gaddes, 1980; Olson, 1984; Restak, 1979), differences in the achievement patterns of boys and girls can be explained on the basis of innate differences in brain functioning. This research suggests that girls perform better in reading and boys in mathe-

3 1 7 Unfair Discrimination, Poverty, and Minority GOed Children 309 matics because of differences in their modes of cerebral functioning (see Harness et al., 1984). Since reading problems cause teachersthe most concern, boys are referred moreoften. The mathematical difficulties that girls encounter may be of less concern to teachers and may notresult in referral or identification.

Environmental factors.According to Good and Brophy (1987), however, differential patterns ofachievement cannot be explained on the basis of innate biological differences. "Performance differences betweenthe sexes are for the most partlearned behaviors (induced by societal expectations and the behaviors of adults)," according to Good andBrophy (1987, p. 32). Their synthesis of theliterature suggests the importance of looking at environmental factors to explaindifferences in the academic performance of girls and boys. The typical behavior of boys and girls in school is conditionedby society. Teachers' expectations tend to mirror more general societal stereotypes: Boys are, and should be, moreaggressive; girls are, and should be, more compliant (Blaubergs,1978; (;ood & Brophy, 1987). Teadwrs' expectations and students' behaviors interact in complicated ways that influence both students' actual achievementand teachers' perceptions about their achievement. According to Carrier (1983, p.963), "docile, industrious, polite, and conforming students ...often get grades higher than their objective performance and aptitude merit."Conversely, aggressive students are perceived more often to be low achievers,and low- achieving students are at risk in school (Miles. 1986). They receiveless positive attention from teachers; their grades, achievement test scores,and even their IQ scores tend todeteriorate over time (Good & Brophy, 1987; Rosenbaum, 1975). These school factors affect special education referral and placement decisions, many of which are made on the basis of students' mistiehavior. In one study (Phipps, 1982), 82 percentof the boys were referred primarily bi..cause of their behavior. Because aggressive behavior influemes teachers' treatment of boys in the classroom,however, such students may, in fact, become underachievers. By the time they are referred, their patternsof underachievement may be entrenched. This reasoning helps explain Clarizio and Phillips' (1986) apparently contradictory finding that gender biasdidnot influence teachers' referral and placement decisions. Both the boys and the girls who wereplaced in 1.1) programs had similar discrepancy scores. Because Clarizio andPhillips did not compare a group of students who were referred with a sampleof students who were not referred, however, they could not report the extent to which girls' underachievenwnt was overlooked.They might have finind, as others have (see e.g., Owen,1978), that underachieving gills are less often referred for testing.

31S $10 Unfair Discrimination, Poverty, and Minority Weer Children The underachievement of girls may go unnoticed because of their compliant behavior in the classroom. Because they get their work doneand please their teachers, girls may be given higher grades than they deserve (Sexton, 1969). Teachers, who typically equate achievement withgrades, may not notice the underachievement ofgirls. Teachers' perceptions and expectations may also reflect society's more pronounced interest inthe achievement of males. Since boys are expected to be achievers, their academic difficulties may cause greater worry than similar difficulties experienced by girls (Schlosser & Algozzine, 1980; Smith, 1983).

Gifted Girls and Women Although women's economic status has improved in recent years. women in the United States still earn less moneythan men. Part of the difference in salaries is due to the history of discrimination against women. Like blacks, their access to administrative positions, even their access to the professions and to business, is more recent than that of white males. Although more women now have positions equal in status to men's, they lack the years of tenure that their male counterparts have. A study of salaries by West Virginia's Department of Education is illustrative. The study found that men hold 66 percent of the highest paying jobs in the state's educational system in spite of the fact that women outnumber men two to One ("More Women itt Education." 1987). The womenwho do hold high-paying positions earn less than men in cc,mparabk positions. Women's relative lack of economic resources becomes apparent when they are divorced from their husbands. In the economic changes that accompany divorce, women's standard of livingdeclines by 73 percent; the husband's rises by 42 percent (Rose, 1986). The fact that the children of divorced parents typically stay with the mother may help explain divorced women's relative poverty. "l'he effects of racial discrimination are cominned with those of sex in the case of black women; and their economic resources are even more limited than those of white women. More than half of the black women who are heads of household earn less than $10,000 per year (Rose. 1986).

Scholastic achievement.Girls tend to earn better grades in elementary school than do boys; but some studies findthat by the time girls are in high school, their grades are no longer higher than boys (Gage & Berliner. 1975). Boys' poorer elementary school classwork may be a function of social pressures to behave in ways that are incompatible with high performance. The negative effects ofdiscrinnnation against girls. on the other hand, are seen in their low mathematics achievement and in the comparatively limited forms their adult achievement takes.

3 1 ;4 Unfair Discrimination, Poverty, and Minority GVtrd Children 311 Law math achievement. The most consistent sex-linked difference researchers have found is in spatial ability (Maccoby & jacklin, 1974), but differences have also been found in mathematics achievement. Although the differences at most levels are trivial, at the higher levels of ability they are dramatic. In the mathematics competitions sponsored by Johns Hopkins University's Study of Mathematically Precocious Youth, girls scored in the highest ranges with much less frequency than boys (Benbow & Stanley, 1981). Twenty-seven percent of the seventh- and eighth-grade boys scored higher than 600 on the SAT-Math; notone of the girls scored higher than 600 (Benbow & Stanley, 1980). The Johns Hopkins findings are mixed, however. It is, for example, interesting that the highest group of girls scored higher on a spatial measure than 1.id the itighestgroup of boys (Astin, 1974). The difference in boys' and girls' math achievcaentseems to increase with age. Astin (1974) found that between the seventh and eighth grades, gifted boys' SAT-Math scores increased by more points than the girls' means. Benl)ow and Stanley (1981) ascribe the differences they have observed to differences in biologically influenced intellectual factors. None- theless, other findings make it difficult to interpret the interaction ofgen- der-related biological differences and the enviromnent. For example, the fact that economically disadvantaged females score higher thaneconom- ically disadvantaged males on both mathematics and verbal tests (Maccoby & .jacklin, 1972) suggests that socialization can strongly influence the character of achievement differences between males and females. It seems quite likely that the differences in the mathematics achievement of males and females can, with equal justice, be attributed to differences in how boys and girls experience mathematics. Girls and boys experience mathematics unequallv. Sells (1976)reports a study in which only 15 percent of the white females and 10 percent of the black femaks had the math prerequisites necessary fora math or science major in college. Casserly (1979) found that girls take fewer Advanced Placement courses than boys in math and science, and she fOund that counselors sometimes discourage girls fi-om taking advanced math classes. Fewer women than men pursue careers in mathematicsor science. Women who do succeed in math may tend to ident4 with their fatlwrs (Eidson, 1971). Several studies report that girls who achieve at high levels in mathematics score high on masculine personality traits (e.g.,Fox & Cohn, 1980). It should be recognized, however, that to referto certain personality traits (like aggressiveness and independence)as masculine may be to conform to outdated stereotypes. Girls' attitudes toward mathematics are most likely conditioned bY complex family Mteractions; however, clear trends do notappear in the 312 Unfair Discrimination, Poverty. and Minority Gifted Children research. Nelson (1971). for example, found that creative female mathematicians tended to be oldest daughters. Astin (1974), on the other hand, found that mathematically talented girls tended to be youngest daughters from large families. As Astin points out, the differences in boys and girls' achievement may, in part, be a difference in educational level of the parents. The girls' parents in Astin's study were not quite so welleducated as the boys' parents. According to Astin (1974) other factors in the family may also play a role in promoting children's interest and achievement in mathematics. During childhood, boys are given toys that are more likely to develop spatial ability than the toys given girls. Boys' parents are also more likely to purchase books, puzzles, science kits, microscopes, and other scientific toys for them. The parents of boys are more likely to provide them with tutor- ing and other instructional activities. Although the parents of all mathematically talented children indicate exceptionally high interest in mathematics and science. the parents of mathematically talented girls are much less apt to show their interest by holding high aspirations for their daughters' accomplishment in these fields.

Achievement and compliance.Much of the antagonism gifted girls elicit from others may result from their failure to conform to social expectations for their behavior. Girls who are bright are sotnetimes viewed less favorably than their male counterparts. In one study, other children used the following negative adjectives to describe gifted girls: "aggressive," "aloof," "bossy," "careless," "conceited," and "dull" (Solano. 1977), Girls who score high on divergent thinking tests may also have greay-r difficulty winning acceptance from other children than do boys who score equally high (Kurtzman, 19(17). Fox (1978) found gifted girls more fearfol of peer rejection than gifted boys. lf, as research seems to indicate, gifted girls tin' more likely to be rejected, girls' fears may be warranted (Solam). 1977).

Adult achiPvement of gifted women. Among Terman's gifted subjects there was a dramatic difference between the scholarly accomplislmwots of males and those of females. Despite equal IQ scores and considerable literary talent among the gifted women, the men in the study publishedhun- dreds of times more books and articles than the women. Sears and Barbee (1977) studied 430 of Terman's female subjects. In response to inquiries about their employment patterns and perceived success, the women who were working outside the home reported greater satisfaction with their lives than other women in the sample. Gifted women who were heads of household expressed particular satisfaction. The resuhs of this study may not be representative, however. As with any self-report study, the accuracy of the subjects' responses influences the study's conclusions. In this study the respondents had sonic compelling reasons to be less

3 Wait. Discrimination, Poverty, and Minority Gifted Children 313 than candid. These unmarried women may have been defensive alxmt their status and, therefOre, may have exaggerated their reported levels of' satisfaction. Other studies suggest that, in general. more women than ineti are disappointed with their lives (Miwcoby & Jack lin, 1974). According to Maccoby and Jack lin (1974, p. 153), women "tend to look back with some regret on what they now see as missed opportunities."

GIFTED STUDENTS IN RURAL SCHOOLS

Recently, educators have been concerned about the quality of schooling in rural and small school districts. The following discussion needs to be interpreted in light of the consistent finding that rural schoolsand small schools in generalprovide a better-integrated educational experience fin students of all backgrounds and achievement levels tkm do large urhan or suburban schools (Barker & Gump, 1964; Stockard & Mayberry, 1986; Tyack, 1974). In fact, some evidence sui!gests that once socioeconomic status is controlled, small schools also produce better achievement among students (Stockard Sc Mayberry, 198(i). This same literature nonethekss suggests that very bright students are at an academic disadvantage in small rural schools. Educators are particularly concerned about the difficulty that such distrkts have in providing adequate academic programs for their most academically able students.

Rural, Small, add Isolated Schools: Definitions Recent interest in the problems of rural education has sparked some discussion almut the distincfion between rural and urban environments (De Young, 1987). One dif ficulty in making this distinction is the variety of geographies, economies, political realities, and cultures that characterize rural districts, Some rural citittns are poor, some affluent; sonw are white, some black; some speak English, some Spanish. Focusing on the range of this variation. however, serves to obscure the long-standing disparities that have always existed between urban and rural districts. For example, wherever they live and whatever pursuits they follow, rural residents do tenti to be poorer than metropolitan residents; they are dose to nature because they work as miners, farmers, and fisher- men; and they are more apt to know all about their neighbors' lives. In order to study common phenomena, then, some definition of ruralness is needed. The National Rural Development Institute (NRD1, 1986) defined a rural area as having fewer than 150 inhabitants per square mile or as being located in a county in which 60 percent or more of the population lives in communities of' 5,000 people or less. They defined "small school district" as one with an enrollnwnt of 2,000 OF less. According 3 2 314 Unfair Discrimination, Poverty, and Minority Gifted Children

to NRDI, an isolated district or school was located at least 100 miles from a district that was not small. Some organizations base their definitions of rural counties on the definition of "urban" used by the Federal Office of Management and Bud- get (OMB). In such cases, everything that is not considered urban is classified as rural. The OMB definition of "urban area" is based on a variety of demographic and economic charkteristics. Basically, an urban area is associated with a population center of 50,000 or more. It has a population density of at least 1,000 persons per square mile and is bound together by an effective infrastructure (i.e., highways, railroads, and communications

Characteristics of Rural Gifted Students In studying rural gifted children, some cautions should be kept in mind. First, rural children are a diverse group. 1 here may be more differences between some groups of rural children than between rural and urban children in general. Low-income rural children, for example, are different from rural children whose families' incomes are average or above. Rural children who are isolated and who belong to cultural minorities are probably different from rural children whose families bekmg to the cultural mainstream. In addition, it is probable that the majority of rural children who have been identified as gifted according to commonly used standards (e.g., achievement or IQ scores in the 95th percentile or above on national norms) are more like middle-class urban and suburban ch.fren than like most rural children. Consequently, the characteristics of identified gifted children may not be very helpful in identifying bright rural children whose academic achievement has been inhibited by lack of support for their education. One significant characteristk of rural students in general is that they spend less time in school than urban children. They spend fewer days in school during the school year; and about 65 percent of rural, low-income children drop out of" high school (Kuvlesky, 1973; Riddel, 1971). These findings suggest that the problem of underachievement among rural gifted students is related to conditions in rural schools. Rural communities may not have the resources to provide stimulating and varied academic programs, and they may not have fimds to offer any special opportunities for their brightest students (Birnbaum, n.d.).

Identification problems.Educators often fail to identifas gifted tlw brightest students from poor rural districts. Such students aiprobably excluded because of the effects of their poverty rather than the effects of their rural upbringing. Because they are not identified as gifted, however.

323 Unfair Discrimination, Poverty, and MinorityGOed Children 315 these bright students may be deprived ofthe educational experiences necessary for their optimalcognitive development. Procedures used to identify brightstudents do not typically result in the identification of many poor, ruralchildren. Even the brightest students in poor rural schools may have IQ orachievement scores that are below the cut-off level fOr admission to gifted programs. Educators sometimes adopt suhjectiveidentification nwthods because they believe that such methods will serve toidentify gifted students from poor, rural backgrounds. Suchmethods, however, may be lessculturally fair than they seem. Teacher nominationsbased on checklists of characteristks of gilled children may locatethose children whose behavior is most like that of middle-class children; they mayfail to identify those rural children who ha1 e the highest level of'academic achievement within their peer group.

Characteristics of Rural families and Communities Cultural values in rural areas may contribute tothe underachievement of the identified giftedstudents in flume areas. In addition, they may create conditions that make itunlikely for many rural students to make high scores on standardized tests of abilityand achievement. Finally, they may condition postsecondaryschool and career decisions that limitthe adult attainment of the brightest studentsfrom rural areas. Because the family is a crucialsocial unh in rural communities, it tends to be the most influential agency forpromoting the cultural values of the community. Church and neighborhood groupsalso serve a socializing function. Though some characteristicsof rural families tend to promote achievement. others tend to limit it. Thefollowing discussion considers some generalizationsabout the character of rural families andrelates these characteristics to the achievement motivationof bright students in rural areas.

Characteristics of rural families.Rural families tend to be less accepting of diversity of' belief andbehavior than urban families (Willits. Beater, & Crider, 1973). Becauserural communities are more insulated than urban communities, their characteris influenced by fewer agents of change. Families in such communitiesoften preserve the traditions of previous generations and resist diverseideas (Birnbaum, n.d.; Photiadis, 1983). Gilled children, however, areoften divergent, and their patterns of development may differ considerablyfrom the norm. Families of such children may attempt to suppress tlwchildren's divergent behaviors and deemphasize their developmentalprecocity. Because they are concerned with preservingtraditio-.. rural families

3 74 316 Unfair Discrimination, Poverty, and Minority Cyked Children tend to cmfiirm more rigidly to sex-role stereotypes than do urban families (Kleinsasser, 1986). I fence, they are quite likely to distinguish the kinds of activities and studies that are appropriate for boys fi-om those that are appropriate for girls. Rural families may also be more concerned with the achievements of boys. Whereas boys are expected to choose a career as well as to have a family, girls are often expected to behomemakers only. These values limit opportunities for both boys and girls. Rural boys may less likely to study subjects such as art and literature, which are considered to be "feminine." Girls may avoid "masculine" subjects like math and science. Additionally, bright girls may choose to marry early rather than to pursue a college degree. Rural families are often ambivalent in their attitude toward education (see Box 11-1). Many rural families perceive tlw school as an unwelcome influence cm their children. In rural communities where religious values are well defined, the school may be viewed as an agentof suspicious secular influence (Cummings, Briggs, & Mercy, 1977). Even when religion is not an issue, parents may consider the school curriculum tobe irrelevant, or even hostile to their values. Finally, some families may view education as a way for the larger society to lure their children away from the rural community (Nachtigal. 1982; Sher, 1977). Families may fear that when students learn new skills, they will develop higher aspirations. They believe that such aspirations will ultimately cause their children to leave the community. Unfortunately, such fears are based on real conditions. Rural ecomnnies often cannot support too many professional, managerial, or scholarly jobs (Nachtigal. 1982; Sher, 1977). "fhose rural students who get advanced training mav need to leave the community in order to find suitable work.

Cha racterist ics of rural communities.'Mc ethos of- the rural community is more egalitarian than meritocratic (Counts. 1930). In such an ethic, it may be considered uncouth to claim special privileges or services, even if some special provisions, like gifted education, seem to makepedagogical sense. This factor may explain both why it is difficult toestablish any pro- grain for apt learners in a rural distriet, and why, whenestablished, such programs tend to enroll disproportionate numbers of the childrenof' the local elite (cf. Mills, 1959, pp. 39-42). This dual observation makes it necessary to distinguish between gifted programs that serve genuinely ntral populations and those that do not. Even in rural areas, gifted programs seem to be most strongly established in the largest towns, an observation that appears to be consistent with the findings of Cox and associates (1985). In many ways gifted children in programs located in these towns do resemble typical gified students.None- theless, even in large rural towns, the egalitarian ethos may be stronger than in affluent suburban or urban districts.

3 9 5 Unfair Discriiptination, Poverty, and Minority Gtlied Children 317

BOX 11-1Case Study of a Gifted Child from Southern Appalachia

In 1967, a residential high school, the Lincoln School, was established incentral Kentucky to provide a program for gifted, but low-income and culturally different students. The students were nominated by local school district superintendents, who based their nominations on the students' socioeconomic status, ethnic or racial status, and potential for achievement. This report is about one of the 240 students who attended the school during the few years of its operation. It is taken from "JuniorGifted and Disadvantaged,' a chapter in Hauck and freehill's (1972) The GiftedCase Stud- ies.

Junior's hometown is an isolated rural community in westernKentucky. The c ommuni- ty , an approximately 800 acre area, hasfive Protestant c hurches and one grocery store, which is also the post office. The population is about 200.The nearest elementary school is fifteen miles away and the high school slightlyfarther. Most of the r in the town, including Junior's father, earn a livingthrough unskilled or semiskilli oor and farming. Most of the women are housewives. Families in the areahold traditional, conservative views on politics and religion. They rarely leave the community.Going to town on Saturday, going to church, and visiting withrelatives make up most of their soc ial I ife. According to Lincoln School personnel, Junior's family is stronglyinterdependent, valu- ing each other's affection and company, but alsofrequently in conflict. !Although the authors of the chapter assert that juniors family members are extremelyinterdependent and hostile, it is possible that stereotypes about rural Appalachianfamilies influenced their perception. They offer no examples that clearly supporttheir view.) Junior's grandmother appears to be a major influence inhis life. He is her favorite grandchild, and she has taken special interest in his development,reading to him and encouraging his interest in books and in religion. Her attention tohis intellectual development contrasts with his parents' lack of commitment to the educationof Junior and his siblings. The teachers and counselor at Lincoln School believe that the impact ofJumor's community and home is seen in his alienation from his peer group at theschool. His strong commitment to religion, his conservative politics, and his ambivalencetoward racial integration set him apart from many of the other students, Juniorlikes the other students, however, and reportedly has become more tolerant of beliefs that differ from his. His teachers report: -He is torn between 'old time religion' and a kind of hedonism; between being close to nature with little responsibilityand few worries and the satisfaction

of intellectual encounter. . . . Rigid and conservative ways of religion, politics, social relationships, and race have given way to questioning since he came to Lincoln School." Although, with an IQ score of 140 on the WISC, his fears seemed unwarranted,Junior is anxious about his ability to succeed academically at the school. Hefears that he is neither smart enough nor self-disciplined enough to compete with the otherbright students there. He is particularly worried about math, the subject in whichhis achievement is lowest. He needs frequent reassurance about hisability to do the work. The teachers find that he is more comfortable when math and science assignments are explained in detail so that he knows precisely what is expected of him. They thinkthat Junior probably misses the easy success he enjoyed inelementary and junior high school. His exceptional intellectual ability allowed him to excel in his ruralschool

rit 318 Unfair Discrimination, Poverty, and Minority Gifted Children

district without much effort; and, not surprisingly, he enjoyedhis success. His teachers believe that his easy success in the past left him unpreparedfor difficult academic work. They are confident that he has the intellectual ability to masterthe material they present. lunior's IQ and achievement scores ondifferent tests vary (as they do with most children); but, with the exception of one math subtest, they areconsistently above average. His group IQ test scores range from 106 IQ in thefirst grade to 127 IQ in the eighth. Although this may represent real differences in his ability atthose age levels, the limitations of group tests often result in gifted children'smaking lower scores than they do on more comprehensive, individually-administered tests.The apparent increase may simply reflect differences in the tests at differentgrade levels. In the eighth grade. junior's grade equivalence scores on the CaliforniaAchievement Test were in the eleventh grade in reading comprehension, in the twelfthgrade in grammar, and in the tenth grade in arithmetic. His total score was in the95th percentile on national norms. On an achievement test given in the ninth grade, Junior'shighest scores were in social studies and literature (both in the 99th percentile); his lowest score wasin quantitative thinking (math), in which he scored in the 40th percentile. junior's anxiety about schoolwork is manifest inhis fear of evaluation and in his sharp self-criticism. His classsroom performance fluctuates fromday to day; however, he excels in English and social studies. l-lis compositions forthese classes are particularly well written and creative. In math and science, he succeedssporadically. In spite of his fear of math, he scored above average (in the 75thpercentile) on a standardized algebra test. Teachers attribute his low grades tohis poor study habits and ambivalent attitude toward school.

Sourte. Adaptvd from W Brown, 0 , Bynum, (& linhInson, S 11972) lunior(AIM and Disad vanugedIn B B. Haut A&MI I rechill dds 7,1 hr ( Atueftr% ipp Bb It/47, Dubuque, IA Brown I Used by permmion.

Characteristics ofruralschools.Althotigh the underachievement of bright studentsinrural areas may relateto family and community characteristics, it may also relate toconditions inrural schools. Rural schools are often located in low-income -treas. Therelme they may be unable to afford the cost of high-quality academk programs or of special teachers for gifted students. The effects of historically underfunded schools are felt bv all the students, hut they are as academically damaging to able students as they are to impoverished students. Lack of appropriate programs may significantly harm the cognitive development of bright students in rural schools (Birnbaum, n.d.). Act ord- ing to Plowman (n.d., p. 73). gifted Cuildren in rural areas arelikely to be "( I) isolated from intellectual stimulation and from lear»ing resources: (2) unsophisticatedunifOrmed, lacking in social and learning skills; and deprived culturally and educationally." Why do rural schools have such a difficult time providing academic programs for the brightest students? Thefollowing list presents some HI the constraints that rural schools encounter. Unfair Discrimination, Poverty, and Minority Gifted Childrrn 319

Rural schools in some areas may not have sufficient space to house special programs for gilled students. 2.Small rural schools may not have sufficient numbers of gilled students to just4 the implementation of special programs fOr these students. 3.The distances between rural schools within one district may be so great that the cost and time of transportation to centralized gifted programs maybe prohibitive. 4.Parents a rural gifted students may oppose programs that requiretheir children to ride buses to gifted programs located at schools other than their local school. 5.Rural high schools may have very kw ekctive options (e.g.. science courses, foreign language courses, and music courses). 6.Rural high schools may he unable to provide advanced-level worws in required subjects such as English and mathematics.

The family, community, and school factors described previously help determine the nature of rural children's schooling in that they affect children's and parents' attitudes toward school and what is taught there. In addition, as Riddu (1971) suggests. the same factors affected many of the children's teachers when they were in school. Although these teachers may be as conscientious and hard-working as urban or suburban teachers,their education often suffered from the same lack of community resources and commitment to intellectual concerns as that of' their students. Because they were not exposed to an excellenteducation system with ample resources and a public committed to high educational standards, rural teachers' expectations regarding student achievement are sometimes far too low. Con- sequently. bright children in rural school districts are seldom challenged to achieve their potential. They are in a "disadvantlged" school system. In some rural states and regions, the longer rural students are inschool, the further their performance falls below that of the national norm.

SUMMARY This chapter reviewed the effects of unfair discrimination on verybright students. The discussion reminded readers thatdiscrimination is an established fact and then proceeded to examine the qualityof that discrimination, in particular the nature of' its impact ondifferent groups. Anumg these groups, low-income students figured prominentlybecause poverty is a condition associatedwith distinctions of social class. The discussion evaluated the underachievement of bright students fromminority groups and from rural areas in light of evidence about theeffects of discrimination. Discrimination associated with gender was also examined. It was noted that under certain circumstances, males may bethe object of unfair discrimination, as in the case of their disproportionaterepresentation in programs intended fOr learningdisabled students. In other cases. lemaks

4,) 320 Unfair Discrimination, Poverty, mid Mixerity Oted Children

arethe object of discrimination. Their achievement in particular subjects such as science and mathematics is often adversely affected, and their career attainment may also be inhibited. In general, discussion suggested that among groups who suffer the effects of unfair discrimination, the most crippling educational effect is diminished expectations. Low expectationsand internalized responses to those expectationscontribute not only to the underachievement of some students but also to their decreased social mobility and limited economic competiveness. 12 Methods of Talent Development

I. Focusing Questions II.Identification Procedures A.Referral 1. Effectiveness and efficiency of referral proce&res 2. Referral sources B. Assessment 1. Intelligence testing 2. Achievement testing 3. Measuring creativity and talent in the arts 4. Unmeasurable characteristics III.Placement Options and Procedures A. Why Do Gifted Students Need Placement Options? B. What Sorts of Placement Options Are There? 1. A continuum of service C. How Do We Determine Placements? 1. Determining gifted students needs D. Is It Really Necessary to Be So Careful? IV.Acceleration A. Disputes About Acceleration 1. Identification 2. Socialization needs of gifted students 3. Higher-order thinking skills B. Types of Acceleration 1. Accelerated regular class placement 2. Special accelerated classes 3. Early entry 4.Dual attendance , 1/tif) 321 322 Methods of Talent Development

5. Other forms of acceleration V. Enrichment A.Process-Oriented Approaches 1. Problem-solving techniques 2. Divergent thinking skills 3. Metacognitive strategies B. Content-Oriented Approaches 1. Course delivery 2. Course content C.Product-Oriented Approaches D. Why Enrichment is Controversial VI.Curriculum: Academics and the Arts A.Academics, the Arts, and Social Reconstruction B. Curriculum, Instruction, and Individual Differences 1. Curriculum and classroom instructional practice 2. Curriculum and individual differences 3. Implications for gifted students C.Curriculum for the Gifted 1. Elementary level 2. Secondary level D.The Arts 1. The arts and artists in society 2. Improvement in school arts programs VII.Characteristics of Instruction for Gifted Students A.Instructional Formats 1. Which format is best? 2.Inductive formats 3. Expository formats 4.Caveat magister (teacher beware) B. Instructional Techniques 1. Teacher as intellectual model 2. Techniques for concept development 3.Techniques for academic skill development VIII.Summary

Focusing Questions 1. How should a school district's identification processrelate to its definition of giftedness? Are some definitions easier to operationalize than others? 2. In what way can the special education "continuumof services" be adapted so that it provides suitable placementoptions for gifted students? 3.Why are accelerative options essential to the educationof gifted students? In what way do these options enablegifted students to develop higher-order thinking skills? 4.What is the distliction between a broad notion ofcurriculum 3 31 Methods of Takost Develop:west 323

enrichment and a narrow one? Which conceptionsupports more appropriate programs for gifted students? 5.Why should academics form the basis ofa curriculum for gifted students? 5.What contributions do inductive, as opposedto expository, formats make to the development of academicconcepts and skills?

IDENTIFICATION PROCEDURES

Before it is possible to provide special educational servicesto gifted students, schools must identify such students. Identificationmethods depend, in large part, on the definition of giftedness thata school or district adopts. lf, for example, a school considers giftednessto be high intelligence and high creativity, the school will needto use measures of intelligence and creativity to locate the most gifted students. If,on the other hand, a school defines giftedness as artistic talent, it will needto develop techniques to identify those students whose artistic performance is themost exceptional. Although this precept seems so obvious that it hardly needsto be mentioned, it is violated nearlyas often as it is followed (Howley et al., 1986). Part of the reason is that not all definitions of giftednessare equally operational. Although it is fairlyeasy to measure academic potential, for instance, it is much more difficult tomeasure leadership potential. Districts that attempt to identify gifted leaders will, therefore,find it difficult to locate technically adequate measures of this characteristic.Similarly, districts that include creativityor social intelligence in their definitions of giftedness will have a hard time finding appropriatemeasurement instruments. Regardless of their definition of giftedness, however,most districts use a similar sequence of identification procedures: referral,assessment, eligibility determination, and placement.

Referral Student referral is the first step in the identificationprocess. A teacher or parent may be the one to make the school aware thata particular student demonstrates exceptional abilities. Often, schools solicitreferrals from teachers. Less often, they refer all students whoscore at a certain high level on the group ability or achievementtests that are administered rou- tinely. In order fbr the referral process to work well, however,teachers or parents must refer those students who are most likelyto be eligible fbr placement in the gifted program. Sometimes this isa difficult determination to make. The patterns of performance thatstudents demonstrate in

3 '4 2 324 Methods of Talent Development classrooms may not reflect their true levels of achievement. On the other hand, students' performance on group aptitude or achievement tests may be more predictive of their performance on the individuallyadministered assessment instruments most commonly used todetermine eligibility. Effectiveness and efficiency of referral procedures. The goal of an identification effort is to locate all of the students who are eligible for placement in a program. The effectiveness of an identification procedure represents the degree to which the procedure is able toselect all of the eligible students. For example, a referral procedure that locates 95 percent of the eligible children in a school is said to be 95 percent effective. At the referral stage, it is desirable to identify a sufficient percentage of potentially eligible students so that no eligible students are overlooked. Many school districts will also be concerned that they avoid referring too many students who turn out not to beeligible for placement in the program. Students who do not qualify areoften disappointed and, erroneously, consider themselves tshave failed. The effkienty of a referral process represents the extent to which the processincludes students who are eligible fbr placement and at the sametime excludes students who are not eligible. In order for an identification procedure tobe optimally effective, it will necessarily be less than optimally efficient. Referral procedures that are 100 percent effective tend to be about 50 percent efficient. Schools that make the commitment to find all of the swdents who will qualify have to test quite a few students who do notqualify. Schools that prefer to avoid excessive or unproductive testing have to o-cept the fact that they will overlook some of their gifted students.

Referral sources.According to Marland (1972), classroom teachers are the most common source of referrals.Nevertheless, classroom wady_ rs may not be the best source of referrals. Teachers maymistakenly assume that the most compliant students in their classes, or the ones with the highest grades, have the highest levels of ability. A much more dlr. (live referral method involves the use of group ability or achievement tests. Pegnato and Birch (1959) compared several methods of referral and concluded that group tests were much more eTfec- live than teacher nomination. The most effective referral method they found involved a dual criterion: a group IQ score of at least 115 (mean = 100, sd = 15) or achievement scores in reading and math that were three years above level. Using this referral method,the authors were able to identify 96.7 percent of the gifted children in their sample. This high degree of effectiveness contrasted sharply with the effectiveness of teacher referral: 45.1 percent. Methods of Talent Development 325

Assessment The second step in the identification process involves the assessment of those students who are referred. Assessment usuallyincludes testing. It may also include the evaluationof students' creative pmducts or performances or the observation ofstudents' behavior. The accuracy of the assessment process is increased wheneducators follow certain guidelines. First, they should always choose assessmenttech- niques that correspond to the districts' definitionof giftedness. If the district defines giftedness as superior academic ability,educators should choose the best techniques for measuring academic ability.Considering the imprecision of grades as a measure of academic performance,their use to identify academically able students is questionable. Usingindividual intelligence tests or achievement testsis much more appropriate. Second, educators should select assessment techniques that havebeen shown to be both reliable and valid. Reliabk measures give the sameresults in repeated administrations. This characteristic of an assessmentinstru- ment is important because it assures that a scoreobtained on one administration of the instrument is likely to resemble a score obtained on asubse- quent administration of the sameinstrument. The more reliable a test, the less error is reflected in a given score. The validity of an assessmentinstru- ment, on the other hand, shows thedegree to which the instrument measures what it purports to measure.This characteristic of assessment instruments is determined in a number of different waysdepending on whether the content of the instrument, its conceptual basis, or its predictiveability is being scrutinized. Finally, educators should be cautious in their interpretations of assessment results. Because assessment instrumentsprovide samples of behavior, they reveal only limited information about students' perfOrmance(Salvia & Ysseldyke. 1987). In addition, a score on any assessment instrument necessarily reflects a certain degree of error. Cautious interpretation of test results includes consideration of the error involved in the administrationof the instrument.

Intelligence testing.Because most definitims of giftedness implicate superior intelligence. IQ tests are among the most common tests given as part of school districts' assessmentprocedures (Alvino, McDonnel. & Richert, 1981). Many different IQ tests are available, and districts vary widely in their choice of tests. Group intelligence tests are popular in some school districtsbecause they are easy and inexpensive to administer and score. Such tests,however, have a fairly broad range of error. Scores obtained onthese tests may

3 $26 Methods of Talent Devi! &pond actually reflect characteristics other than academicpotential. Because of their format, such tests may really measure achievementin reading comprehension rather than academic potential. Some veryintelligent children who are not good readers or who process informationslowly will be penalized by group tests. Individually administered screening tests are alsoused by some schools. These tests measure very limited samples ofbehavior: they are neither sufficiently reliable nor sufficiently valid to use asthe basis for decisions about students' eligibility. The best intelligence tests available are individuallyadministered evaluation instruments such as the Stanford-Binet IntelligenceScale and the Wechsler Intelligence Scale for Children-Revised(WISC-R). Such tests take approximately two hours to give and can beadministered and interpreted only by a psychologist or psychometrist. Nevertheless,using this type of test is warranted. Theseinstruments are tnore adequate technically than other measures of intelligence, they sample awider variety of behaviors, and they are better able to predictstudents' academic perfOr- mance.

Achievement testing.Definitions of giftedness that include academic talent require that students' achievement be evaluated todetermine eligibility for placement in the gifted program. Students'academic achievement may also be measured as a wayof determining their levels of intelligence. Since superior intelligence represents the potentialfbr advanced academic work, high academic achievement implies highintelligence. Students whose academic achievement is in the superior range do havethe intelligence necessary to achieve atsuch high levels, regardless ol their actual scores on tests that purport to measureintelligence. As with intelligence tests, there are three kinds ofachievement tests: group tests, individual screening tests,and individual evaluation tests. Again, the individual evaluation instruments are the ones mostuseful for guiding placement decisions. This observation relates particularly tothe measurement of elenwntary schoolchildren's achievement. To evaluate the achievement of older students, educators may need torely on the use of off-level group achievement tests (Stanley, 1976b). Thistesting practice. while not optimal, is necessary nonetheless: There are noindividual evaluation tests that adequately measure the athievement ofolder gifted students. Measuring creativity and talent in the arts.Some districts include creativity in their definitions of giftedness. Hence,educators must use instruments that intend to measure creativity. Such tests areof limited value, however. The relationship between high scores onsuch tests and Methods of Talent Development 327 superior creative performance is speculative. Many tests of creativity actually measure characteristics such as ideational fluency or intelligence (Hocevar, 1980). In addition, creativity tests have such poor reliability and validity that their use to guide placement decisions is unwise. A better approach is for districts to measure creative performance in the arts rather than to attempt to evaluate general creative potentud. To determine arts talent, however, is both difficult and costly. Districts with well developed programs fbr the artistically gifted usually rely on panels of- experts to judge the artistic talents of those studentsreferred for assessment (see e.g.. Chetelat, 1981). Unineastireable characteristics. Some definitions of giftedness include characteristics that cannot be measured. Leadership potential and stxial intelligence are among such characteristics. When districts include these characteristif:s in their definitions, they most often rely on the judgment of teachers to identify those students who are eligiblefor placement. Basing program eligibility on such subjective assessments is troublesome. Teachers are not the best judges of students' academic aptitude. Can we assume that they are better judges of leadership potential orsocial intelligence? The relevance of such characteristics is also questionable. Not much evidence supports the claim that leadership in the classroom predicts political or intellectual leadership in adulthood. Districts can probably strengthen their operational definitions of giftednessby eliminating such categories.

PLACEMENT OPTIONS AND PROCEDURES Developing placement options and procedures is less -technical- than determining reasonable identification methods, but it is not necessarily easier. In the case of placement. educators and parents must understand why gifted students, as opposed to other students, can benefit more from certain instructional environments than from others. They also need to know that many people--including some educatorsdo not share this understanding. Therefore, it is important for those who advocate gifted education to define clearly the options and procedures that entitle gifted students to special programs.

Why Do Gifted Students Need Placement Options? Verv bright children encounter academic difficulties in the unmodified regular classroom (Lewis & kanes, 1980). This is not surprising. According to the assumptions behind the high-IQ definition of giftedness, 328 Mothods of Talent Develop Rad gifted children in a typical classroom are moreadvanced than all of the other children. They are more rare than 1 in 30students. Because there are so few high-IQ studentsin the regular classroom, teachers cannot afford to spend too much time cultivatingthe academic talents of such children. It would not be efficient for them to do so.'therefore, in most cases, the regular classroomis no- an environment of particularbenefit to gifted students. Before turning to a discussion of environnwntsthat better accommodate the needs of gifted students, it mightbe wise to consider a popular argument that has beenadvanced in deffnm of regular class placement for gifted students. This argument (e.g., Henson,1976, p. 110) runs as follows:

Gifted children will need to interact with averagepeople all their lives. Gifted children, after all, are destined to become our nationalleaders. 11 they do not associate with ordinary people. they will become sociallyisolatki. As a result, they will not be able to lead. They will have losttouch with the concerns of average people.

This argument has a wmpdling emotional appeal.At the same time, however, it is based on someshaky assumptions. Of these shaky assumptions. one is, ironically, elitist. First of all, gifted children are not "supernormal."In many respects they are average. They are, for example,hard to distinguish from average students on the basis of social adjustment,social distance, or social attractiveness (Maddux, Scheiber, & Bass, 1982). Theytend, pethaps, to be a bit more well adjusted than averagestudents when they are put in accelerated placements (Braga, 1971; Robinson & Janos, 1986;Kulik & Kulik, 1984) but the social difTerences are never very greatin either negative or positive directions. Second, though many people are fond ofsaying so, there is little evidence to suggest that gifted students are any morelikely to become leaders than other students of the samesocial class. 'terman's work supports this finding ("Ferman &Oden. 1959). In liollingworth's (1942) view, the highly gifted were le.s.s attractive as leadersthan were students of only moderate ability. 'third, to promote the gifted as "natural" leadersis elitist (Weiler. 1978). It is odd that such undemocratic principlesand values are espoused so openly by someeducators (and some parents). After all, leadership demands a variety of abilities, and academicability is only a small part of what is required for leadership (Bass, 1981).Many gifted students will demonstrate only average interest in and ability tolead. In fact, leadership within a professionwhich is where we might expectit to be demonstrated in most casesis nut strongly related toacademic ability or achievement (Rail-d, 1985). Methods of Talent Development 329

What Sorts of Placement Options Are There' There are enough gifted children in any school or anyschool district to provide a pool of students for agifted program. It is clear that these children do not do nearly so well as they might because ofthe limits imposed by their placement in the regular classrixmi. It isequally clear that something of benefit ought to be arranged for thesestudents. The discussion alxwe, however, suggests that the regularclassroom is not an environment thatprovides appropriate academic expectations for gifted students. One way to get an overview of moresuitable placement options is through the notion of "continuum of services," a conceptthat is familiar to special education teachers. Howley and associates (1986) adapted thisnotion to include gifted students. The organizing principle of the original continuumfor handicapped students was wial involvement. The organizingprinciple of the adaptation for gifted students is academic advancement.Academic advancement, of course, reflects the unconmionacademic talent and rapid learning rate characteristic of gifted children.The continuum Of services in this interpretation can also accommodate the academic delayof some handicapped students.

A continuum of service.In this adaptation, the regular class represents the mid-point of the continuum of services.Here is the continuum fOr gifted students:

special sclumladvanced program (7no,/ advanurd) Nprcial classadvanced program resource programadvanced program regular classdevelopmental placenwnt regular classemiched program regular classregular program (loot uth c(1)

The basic point of using such a continuum is to illustrate the fact thin a vurwtv of placement options is neeessarVin order to accornnuxfate the various needs and talents of the gifted students within any district.A district's program may, of necessity, emphasize one option oranother. Usually, developmental placement. resource program, andspecial class placements need to be the most commonly offeredoptions. Butall Options should be potentially available, to serve students whoseeducational needs cannot be met in placement options that arc actuallyavailable. Among the options that fall within the alxwe continuum arethe fi)1- lowing (ordered from least advanced to most advanced): 330 Methods of Takla Development

Regular ClassRegular Program A gifted studem may he left full time in the regular classroom.This placement should be made only if a gifted studentexhibits a serious dist repancv between IQ and achievement in all areas; thatis. if the gifted student's achievement is normal. It may also be the wisest placement if thegifted program is disorganized. is not academicallybased. or of fers to better alternative. Regular ClassEnriched Program A gifted student may be left hull time in the regulardassrocmi with the regular teacher (or visiting teacher of the gifted) providing grade-level lc- tivities On topics related to. but not included in. the regular program. "Ihe goal here is to interest the child, not actually to provide advancedMstruc non. Enriched programs itt resource rooms espouse the same goal and chi lerlittle from this placement. Regular ClassDevelopmental Placement A gifted student mav be given instruction in a higher grade orin a more advanced class in a high-school sequence. This placement may helull time (a- part time. This sort f /I placement is a veryeconomkal wa y. to impkinent acceleration. Resource ClassAdvanced Placement A gifted student may attend classes with the teacher of thegifted to receive advanced instruction. Resounr rooms, however. typkally primideother sorts of services besides advanced placement. and manv resourceteachers travel between several schools. These linfuations can cause problems thatunder- mine consistent instruction. Special ClassAdvanced Program A gifted student may attend a special class. At the elementarv level,the teacher of t he gifted might offer regularly schedukd accelerated reading and math classes. At the high-school level, a school might develop honorssections. or accelerated curricula (e.g.. three yearsof mathematics in two). This opticm might also involve students in taking advanced courses at the college level. Special SchoolAdvanced Program A gifted student might be sent to a special school that offers advancedand rapidly paced classes. A number of such schools exist, especially lot advanced high-school and junior high-school students. In addition, a number of early- entry college programs seek specifically toaccommodate gifted .junim and seniot high-school students. Such programs, too, fall in this categot v.

How Do We Determine Placements? he need to keep placements potentially available challengeseduca- tors' problem-solving skills. In order to organize theplacement of students in particular options, some set of procedures is necessary. Inprevious times, placement was much simpleras was identification. Aneducator selected a test, gave the test to whomever was thought appropriate,and placed students in programs. In fact, students in regularclassrooms are often assigned to particular teachers, programs, or sections in justthis way. Gifted children, however, are in many states considered to he "exceptional students." In fact, the Council for Exceptional Children (CEC)has included gifted students, together with handicapped students. inthis Methods of Talent Devetoptoesd 331 category since 1922. Studies suggestthat gifted education programs are stronger in states that include giftedstudents under the rubric of "exceptional students" (C. Howley, 1986). l'Itere is abig difference, however. between haying somewhat stronger programs at the statelevel and having good placement options and procedures at the locallevel. Strong state leadership is a good sign, but whether or not it leads to goodplacements is unclear. Hence, local districtswhether they have strong stateleadership or notare wise to develop placenwntprocedures for gifted students. The authors bias favors administration of gifted education withinspecial education. When gifted education is administered inthis way, special regulations and due process rights protect the interests ofgifted students, just as they do those of handicapped students. Theseprocedures and rights will not be discussed in detail below: however, oneimportant procedural point must be made. That point concernsthe basis on which placement is made.

Determining gifted students' nerds.'1' he preceding discussion focused on placement options.Placement, however, is the outcome of' a process of deliberation and reflection that begins with referral.Actually deciding where to place a student is the final event in this process.This point cannot be stressed too much. It is critically important not toplace a student until a ter the student's"educational needs" have been determined. Important distinctions exist between the diagnosticinformation tallied in an evaluation, a student's needs, "prescriptions" to meet til:.se needs, and placement. The purpose of evaluation is to provideinformation that can help educators and parents make thosedistinctions. If, for example. a student's IQ is above the 99thpercentile, but the same student's written languageis at the 31th percentile. then, most likely, there exists a need. The nature of the need. however, is a matterof opinion. In this case, the student's need might be expressed in any oneof the f011owing ways, depending on circumstance (e.g...judgment of participants, negotiatitms of the placement nmunittee, and other infOnnationabout the student):

-SLISM1 itt-Cds It) iillprOVC het petionname on the Woodeoti-johnstmWi len Language Cluster bs 0.5 standard deviatnms.-

"Susan nerds to improve her writwn language skillssuhstanti;Als.-

"Susan needs to improve her skill in distinguishing writtenniisspellings (loin correct spellings.-

3 4 ) 332 Methods of Takot Developffserat Once a need is determinedand probably no more than 3 or 4 needs should be defined fOr a particular studenteducators and parents should consider whatimtruction(not what placement!) can meet those needs. Per- haps, in the above case. Susan should work twice a week with a peer tutor who will help her write and edit short essays. Perhaps she should participate in a remedial spelling program (not actually a prescription the authors would commend to anyone). Perhaps Susan should work on a word processing program to achieve successively better drafts of regularly assigned work. All of the above are examples of prescriptions. We have so far said nothing of placement. Once the prescripfions are ready, then placement can be determined. Are the prescriptions best accommodated by the resource teacher? In an available smial class? Are they so unusual that "potentially available" placements must be brought to bear? The reason for waiting to decide placement last is that students needs ought to come before the convenience of the educational bureaucracy. It is mo easy for educators to say, "Ileres our gifted program; let's put this kid there.- without ever reflivingabout what the student is like. Yet, this st e- nario is not at all unusual. If placement occurs within the special education process. then the above sequence of events will help make better placement meetings. That sequence of events provides an agenda for the meeting (interpreting diagnostic information, defining needs. specifying prescriptions, and making the placement); and it involves everyone in the outcome. When everyone has participated in this waY, parents, teachers, and administrators are more likely to support the placement and implement it diligemly. Subsequent interactions among them are more likely to be positive and productive.

Is It leally Necessary to Be So Careful? Teachers of the gifted are dealingor ought to be dealingwith the nwst able students in our schools. The education of these students needs to be taken seriouslv. If placemetn is only a way to enhaiwe students' status and a source of pride for their parents, and programs do not result in more rapid learning and better thinking and reasoning skills among gified students. then we have done an injustice to the students identified as gifted. Moreover, we have done an injustice to those students who were labeled "non-gifted"; these students could have been as well served by the program as the gifted students. One (if the things we can do to help avoid these results is to be more caret ul about how we place students.

ACCELERATION

Because of tlw way in which giftedness is usuJly identified (i.e., high IQ). one of the most tunable characteristics of gil ted t hildren is the accderated

3.1 Methods of Talent Devekposent 333 rate at which they learn. They learn rapidly even when they remain in unmodified regular classrooms. Even underachieving gifted students served in an enrichment program can make two years of academic progress in one year (Fearn. 1982). Accelerated learning is a characteristic of' gifted students. It is also a strategy fin accommodating their progress through school. The discussion above noted that instruction should relate logically to evaluation. flow and what should be taught to certain types of students ought to be determined on the basis of' the students' characteristic's. Acceleration is the strawgv of choice when students are identified as gifted on the ha: is of- unconunonly high scores on IQ or achievement tests. Disputes about Acceleration Not everyone thinks that acceleration is a good idea for gifted students. Many educators involved with gifted students believe that a "qualitatively different" curriculum. not a faster pace, best meets the needs of gifted students in general (e.g.. Castiglione, 1984; Renzulli, 1977; Sisk & BierIy, 1979). Several critical issues are involved in this dispute. Identifica- tion, socialization needs of gifted students, and higher-order thinking skills all play a role in the ongoing contrmersv. Identification. We know that the use of standardized tests discrimi- nates against students from impoverished backgrounds. Some policy options can combat this abuse, but they have not been implemented widely (Council of State Direciors of" (ifted Programs. 1985; Howley. 1987a). Many educators seem to feel that a dif ferent approach is wiser. perhaps because it seems to of ler totally new possibilities for equality. 'I 'hese educators feel that in identifying different sorts cif talent, gifted education will become more accessible to stlidents from the oppressed classes of society. Fhere is, however, little evidence to support this sentiment. In fact, many observers find that programs based on the premises of this approach to identification arc not appropriate for academically able students (e.g., Copley,. 1961; C. Howley, 198(5; Stanley, I 976a; Weiler, 1978). Some find such programs to be elitist and anti-intellectual (Run. 1985; A. I lowlev. 198(5: Weiler, 1978).

Socialization needs of gsfted students.The clevelopment of accelei- ated programsas opposed to the characteristic of accelerated learning within studentsfills some educators and parents with fear. Most children etwounter social problems in school, just as most students encounter academic problems. All educators and parents know that social pecking orders dominate sch,ioi life. Those who are karful of acceleration worry that accelerated placements will jeopardize students' happiness. This point is impo. tant because our culture values happiness very strongly, more strongly, in fact, than learnedness, bea:ity, or meaning 342 334 Methods of Talent Development (Arendt, 1981; Hofstadter. 1963). We value achievement, not because it enacts beauty or meaning, but because it pronwteshappiness. Acceleration appears to bean immrdthtethreat to students' happiness. Hence, many educators and parents oppose it. They fear that gifted students will become social outcasts or, worse, intellectual introverts incapable of sustaining happy relationships with others. These parents and educators need to be reassured. When accelerated and nonaccelerated groups are compared, only minor differences in social and emotional characteristics are found. On the other hand, thereare sigMficant differences in the achievement of accelerated and nonaccelerated groups of gifted students. We know that accelerated college students tend to win more honors, earn higher grades, and participate more in extracurricular activities than equally bright students who are not accelerated (c1. Fund for the Advancement of Education, 1957; Robinson & Janos.1986; Pressey. 1967; Stanley. 1985, 1986).Results with elementary and high-school students are equally positive (Alexander & Skinner, 1980; Braga. 1971; Daurio, 1979;Kulik & Kulik,1984). A highly readable presentation (see Box 12-1) of the issue that can help parents and educators

BOX 12-1Accelerated Enrothmal Development?

Studies indicate ... that while accelerated students are making rapid academk progress, their maturity is speeding up in other ways as well. According toresearch by Julian

Stanley. .accelerated students tend to date earlier, insist on greater freedom, and accept more responsibility than is customary for students in their age groups. The earlier social and emotional development is to be expected. What's more, it's natural, normal, and healthy. After all, these students not only learn to behave like the older students they spend most of their time with, but they also develop a clearer sense that others appreciate their special skills and abilities. Adults shouldn't expect these accelerated students to behave like the children of the same ages who are in lower

grades. Research .. . suggeststhat most of the accelerated students need no special protection and, in fact, cannot be protected from the influences of their more mature environment. At the same time it is foolish to maintain that concern about accelerated students' social and emotional development is never justified. If an acceleration program is not implemented with great care, it could be harmful even for students whose placement in the program is most appropriate. In addition, parents and teachers should remember that, even in carefully run programs, someaccelerated placements will not work out. Recent work by Nancy Robinson demonstrates that a small proportion of accelerated students fail to thrive in an accelerated placement. At the same time, it is not clear that these unsuccessfully accelerated students were harmed, or that they would have been better off left alone. (adapted from Howley, 1987b, p.

Soune Repnnted with permission from Amerman School Board lournal, June, copyright (1987), th National School %larch Mstic tattoo Ali righK reserveci 343 Mdhods of Talent Development 335 consider all sides of the issue reasonably canbe found in the June, 1987, issue of The American Schnol BuizrdJuuriud(Hawley. 1987b, pp. 31-33, 40).

Higher-order thinking skills.Most educators and parents are concerned that our schools are not teaching higher-orderthinking skills very well (e.g., Arends Ifc Kirkpatrick, 1987). Some educators believe that gifted students should receive special instruction in higher-order thinking (e.g., Clark, 1986; Maker, 1982), This line of reasoning sometimes leads to the proposal that gifted students take a course inhigher-order thinking. Since many educators and parents are worried aboutpossible (or imaginary) threats to a child's happiness that might resultfrom acceleration, the proposal to teach an important skill to giftedstudents without acceleration looks appealing on the surface. Educators and parentsshould, however, be aware that such proposals work against the realneeds of gifted students. Mom educators concerned with gifted children now agreethat good thinking and reasoning skills comprise partof academic instruction ((;allagher. 1985; 1 lowley et al., 1986; Swassing, 1985;Van Tassel-Baska. 1985). Itis probable that good instruction at advancedlevels cultivates advanced habits of thinking and reasoning.

Types of Acceleration Grade skipping is the fOrm of acceleration that comes mostreadily to mind. Though itis still not practiced as frey-ntly as is werranteditis probably used more than it was in the period1950-1980. Vet there are at least a dozen hams of acceleration.Educators concerned with gifted students need to be veryfamiliar with all of these forms. 'nes, are the tools with which the "lock stepcurriculum" can be adjusted to the learning raw of gifted students. Some ofthese forms are introduced subsequently.

Accelerated regular class placement.ln this form of acceleration. a gifted student should be allowed to progress throughinstructional materials in tlw self-contained dementarv classroom atapproximately twice the rate of average students. Thisoption requires a well-motivated student. because the regular teacher will not have much time tosupervise the progress of the student. However,this form of acceleration is really a stopgap measure to use when otherforms are not available or not appropriate. Gains made under this arrangement are hard todocument and to maintain as students progressthrough the grades.

Special accelerated classes.In this form of acceleration, a student can attend rapid-paced coursesin academic subjects. Such courses are designed for taudents who can make rapid progress. (Of course. notall students who attend the class need be identified asgifted; students can be 344 336 Methods of Talent Development selected on the basis of advanced achievementin the subject.) Accelerated classesas opposed to regular classesmaintain veryhigh expectatims for student progress. This difference hasbeen identified as a critical influence in such courses. Early entry.In this form of* acceleration, a studentbegins ekmentary school, middle school,junior high school, senior high school, or college a year or moreearly.This form of acceleration is really a specialform of grade skipping. The advantage of this form ofacceleration is that it exploits the disjunctions of the educational system tothe advantage ofgifted students. Research on this form of accelerationis extensive and positive. Dual attendance.In this form of acceleration, a student attends two institutions at once. In the ideal case, a student will receive dual credit for work done at the higher of the two levels. For example, successful completion of' a one-semester miege English course should count also as success- fill completion of a one-year high school English course. This can be useful to students, whose financial aid status in college maydepend on receiving a high school diploma. Other forms of acceleration.Other options that can be used to accomplish acceleration include combined grades,ungraded elementary schools, private schooling, and home schooling. Anoverlooked strategy involves transfer to a different school. Sometimes, when oneschool has a policy that prohibits acceleration or is very restricade in its useof acceleration, a nearby school may better accommodatethe academic needs of gifted students. (See Box 12-2 for a personal account of' the useof ac('eleration.)

BOX 12-2The Experience of Acceleration: A Personal Report

Between the three of us, we have fivechildren. We have used some form of acceleration with each of them. Our five children are verydifferent from one another, and not all of them have received "gifted services." Yet,acceleration has helped meet their educa- t:onal needs. In each case, we think theacceleration was beneficial. Two of us reported in some detail about ourthree children's experiences up through 1984 (Howley & Howley, 1985). Sincethat time, our eldest child has progressed rapidly through school. As a result of her earlier experienceswith acceleration and with dock and a variety of social settings,she seemed to develop her own educational time a sense of her ownacademic values, needs, and expectations. We were a bit surprised, however, when towardthe end of the ninth grade she told us she wanted to enter college in the fall. Weboth worked at the college where she would enroll, and perhaps that made arrangements easierthan they would have been for other students. However, the college had enrolled juniorhigh-school students as fully ma- triculated students before. Some staff membersthere were aware that it could be done, and other parents have made similar arrangementswith the college since that time. At the age of 15, our daughter has nowcompleted degree requirements in math and science (her "weak" areas), and she is majoringin psychology. She will shortly transfer 34 5 Methods of Talent Development 337 to a larger school, where she will liveinthe dorms. She is mature for her age in every way. She conforms to the compositeinlage of successful earl..Itrants. Her grades are very good; she has an active sociallife; she participates 1.. .fiusic and drama at the college and in the community. She is not at all a perfect student, andit may surprisereadersshe is not highly gifted. She does possess the family commitment to reading and music,and to reflecting on social issues. Her success, if that's what it can be called, is aresult of active parenting, her own temperament and proclivities, and hard work.There have been tears, anxious moments, and unpleasant surprisesasthere are between any child and parent. Can we be certain our daughter will lead a well-adjusted,productive lifer No, we can't. We've taken some unusual risks, and we hope forthe best. Matters are pretty much in her hands, however, as she really is more a youngadult than a teenager. That's fine with her. (We're fairly certain she knows that she still has a lot tolearn.)

ENRICHMENT

In contrast to acceleration, enrichment is anoption for gifted students that is not often considered to be controversial.Nonetheless, enrichment programs vary greatly, and someof the types of instruction provided under the auspices of enrichment are indeedcontroversial (Cox et al., 1985).Most people, however, endorse the notion that giftedstudentsshould receive instructitmthat broadens their horizons. Part of the difficulty in evaluating enrichment as analternative fOr gifted students is the fact that so many difTerentpractices are thought to be enriching. Definitions of enrichment reflect theconfusion. For example, some authors considerenrichment in its broadest application. According to Gowan and Demos (1964, p.138),enrichment "embraces all the curricular adjustments made for gifted students. Theseinclude experiences in breadth and depth, ;;nd those connectedwith grouping,acceleration, and individualiEation of iwtructionin heterogeneous classes." Other authors accept a more narrow definition. Epstein(1979), for instance, defines horizontal enrichment as lateralexpansion of the curriculum. She characterizes enrichment as instructionin which:

.the gifted children do not move up any faster, but theirlearning is enriched by different curriculum materials, individual instruction,emphasis on the higher mental processes of divergent thinking andcreativity, independent study, problem-solving projects with real products,and any number of techniques that give youngsters a welconw change from theroutine of the regular classroom. (Epstein,1979, p.69) In order to gain a greater understanding of enrichmentoptions for gifted students, it is important to evaluate severalof the instructional alternatives that are typically used in enrichment programs.These alternatives represent three difTerent approaches toenrichment: process-oriented approaches, content-oriented app.- oaches, andproduct-oriented approaches. 34 338 Methods of Talent Development Process-Oriented Approaches Many enrichment programs are directedtoward the goal of developing students' higher-order thinking skills. Theseskills may include problem-solving techniques, divergent thinking skills, ormetacognitive strategies. Many enrichment programs attempt todevelop higher-order thinking skills in isolation from academic or artistic cmtent.The rationale for this approach is that, once students areproficient in these skills, they will be able to apply them to all content areas(see e.g., Clark, 1986). This approach has been criticized by a number ofeducators (e.g., Copley. 1961; Mariano 84 Hutchins, 1985; Van assel-Baska,1985) who maintain that higher-order thinking skills must be taught within acurriculum based on content in the disciplines.

Problem-solving techniques.Many enrichnwnt programs offer instruction in problem-solvMg techniques. Amongthe teclmiques that purport to increase higher-orderthinking skills are the following:

I.library research projects. 2.problem-solving sessions, 3.Mstruction about the scientific method, 4.future problem-solving competitions. and 5.practice in games of strategy. such as chess andbackgammon.

Divergent thinking skills.Instruction in divergent thinking skills differs from instruction in problem-solvingbecause it emphasizes the significance of muhiple, creativesolutions to a problem rather than the significance of one, correctsolution. Several techniques have been developedfor teaching divergent thinking. Most prominent amongthese techniques are creative problem solving (Osb)rn, 1953) and synectics(Gordon, 19(31). Although these techniques do enable students tobe more fluent in their generation of ideas, they may have noinfluence on students' ability to produce original theories or works of art. Certainlythe impact of such training is seriously compromised when students are notexposed to content in mathematics, science,and the arts. Without a comprehensive Limier- standing of content in these fields. students cannotpossibly suggest creative solutions to rekvant probknis.

Metacognitive strategies.Several educational psychologists(e.g.. Newell 8c Simon, 1972; Sternberg. 1977) havebegun to identify information processing strategies that supposedlyundergird all learning. Accord- ing to some psychologists and educators,students can improve their problem-solving skills by becoming aware of thesemetacognitive strategies and their use. 317 Method, of Talent Development 339

At this point in time, however, research into metacognitive strategies is formative and inconclusive, and programs based on such research are necessarily experimental (see e.g Feuerstein et al.. 1981). '1-he recommendation that gifted programs teach these strategies should probably be viewed as precipitous.

Content-Oriented Approaches Enrichment programs that emphasize academic content rather than cognitive processes are popular in some districts. Such programs offer academic courses that are not typically included in the regular curriculum. The format for the delivery of such courses varies, but the nature of the courses is Often similar.

Course delivery.Some content-oriented enrichment programs enable gifted students to participate in academic courses that are not offered during the regular school day. After-school, Saturday, and summer work- shops can provide academic instruction in subjects such as computer science, .stronomy, archeology, and art. Although these learning experiences may be valuable for many gif ted students, they rarely provide an adequate substitute fnr relevant academic instruction delivered through the regular curriculum. Enrichment courses do not constitute an articulated curriculum: they almost never carry acadennu credit. For these reasons, Stanley (1981, p. 26:4) regards such enrichment as "both the best short term and one of the worst long term approaches to the education of the gifted."

Course content.Often. minicourses are used as the vehicle for deliv- erMg academk enrichnwin. Such minicourses may be topical. thematic, in- skill-oriented. opical minicourses tend to be based on the traditional disd- plMes. An example of a topical minicourse in mathematics would be a summer math course in elementary stiuistk-s. Thematk mthid .rses provide an interdisciplinary view of a particular issue. A course that explores the issue of air pollution from the perspectives of chemistry, biology, aml sociology exemplifies sudi an of fering. Skill-oriented minicourses provide intensive instruction in skills that enhance academic learning. These courses may teach skills such as typing, library research, or test taking.

Product-Oriented Approaches Product-oriented enrichment programs direct students toward the production of significant products. Most notable among these approaches is Renzulli's (1977) Enrichment Triad model. The ultimate goal of this program is to enable students to prepare real products for presentatim to appropriate audiences. Such programs rely heavily on independent study and research. 34 s 340 Methods of Talent Development Some enrichment programs emphasize products that are less tangible. Programs that focus on career education, affective activities, and leadership training have as their products certain behavioral outcomes. In such programs students are expected to mimic the constellations of behavior that are assumed to characterize gifted leaders in business, science, and the profession.s.

Why Enrichment is Controversial Enrichment ü controversial fin- a number of reasons. Because it rarely can address the most significant academic needs of giftedstudents, its value as the sole option for such students is questionable. Becauseit may be appropriate for many average students as well as for gilled students, its exclusive province in .he gifted program can be questioned. The resolution of these controversies seems to be in the return to a broad definition of enrichment. When enrichment encompasses the total program for gifted students, including acceleration, grouping, and relevant instructional methods, it no longer is a controversial provision. Such a broad conception of enrichment, however, might better be referred to as -curriculum." The description of this sort of curriculum for gifted students is considered in the next section of the chapter.

CURRICULUM: ACADEMICS AND THE ARTS

Educators often disagree on what constitutes the best school curriculum. We think that the curriculum for gifted students should be based on academic and artistic di.sciplinek. Not everyone shares our view. Very liberal educators believe that the curriculum can and should include everything that life has to teach. They believe that any other view is narrow and harmful to children (e.g.. Clark, 1986). This view is appealing to many who understand the shortcomings of formal schooling. Very conservative educators, on the other hand. believe that some topics in the academic curriculum undermine traditional values (e.g.. Lines, 1986). Others thinkthat teaching the values of the Protestant ethic is an overlooked priority (Bennett, 1986). Still otherobserversincluding the authorsbelieve that teaching intellectual skills, rather than all life has to offer, should take priority over any other instruction in any value system (Katz, 1971). Advocates of this view understand that some values adhere to any act of teaching and learning. The kind of values that adhere to the teaching of intellectual skills and knowledge are the generic values of hard work, mutual respect of teacher and student, and appreciation of meaningful statements.

3 -* Methods of Taknt Development 341 Children, we believe, must become skilled readers and thinkers. It is an article of faith with us that children who accomplish these goalswill become good citizensin the best sense of the phraseand will inevitably shape their own values.

Academics, the Arts, and Social Reconstruction Schools have shown that their effectiveness is more limited than the great reformers of the pasteducators like John Dewey, George Counts, and Leta Hollingworthwould have liked ( Jencks et al 1979). These educational reformers believed the schools could "reconstruct" a flawed industrial social order (Cremin, 1961: Tyack, 1974). Nonetheless, it seems clear that the schools dare not make a new social order (cf. Counts, 1932). In fact, they cannot At the same time, if they supply students with good intellectual tools, schools can contribute indirectly to the reconstruction of society in accordance with democratic ideals. To do so, they must nurture students who understand history, literature, mathematics, and science. The arts also play an important role in this endeavor by supplyingalternative images of reality in music, dance, drama, painting, and sculpture (Arendt, 1981; Mar- ruse, 1978). If however, schools produce students who know little, read little, and never examine their assumptions, then they will fail to contribute tosocial progress. Ii schools cultivate only the politkally safedisciplines of math and science, they will also fail to contribute meaningfully to social progress. The observations presented here apply to all children, not just to an elite group. More children must have access to the tools of reason. Schools should teach basic skills, of course, but they also should enable students to engage in more sophisticated learning. That gifted children are among those who can make rapid progress through the academic curriculum is a fact that has been confirmed by a large body of research. In our view, it makes sense to accommodate these students' confirmed potential.

Curriculum, Instruction, and Individual Differences Before taking a brief look at what things the gifted ought to learn and when, we need to develop some sense of how curriculum relates to teachers and to students. First, we distinguish between curriculum and classroom instructional practice. Second, we examine the relationship of individual differences in students to both the curriculum and to instructional practice. This discussion applies to all children. Inn it has particular imphcations fOr gifted children. 342 Methods of Talent Development Curriculum and classroom instructional practice.A curriculum is a course of study meant to generalize abody of knowledge to be taught to a given student population. In our view, the curriculum must be anorganiza- tion of disciplined knowledge. (Some handicapped students requireanother sort of curriculum, however.) Disciplinedknowledge is traditional academic knowledge. especially as conceived by those who practicethe discipline. The curriculumis one sourceprobably the most important sourceof classroom practice. If instruction does not contain anadequate intetpretation of disciplined knowledge. then students fail toacquire the intellectual tools they need. The tools of instructioninstructionalformats and techniquesare the means by which teachers interpretdisciplined knowledge to students. Teachers have a choice of instructionalformats and techniques. and they are free to improvise their own. To teachcomposi- tion, for example, a teacher may develop a program of culturaljournalism (Wigginton, 1985), or assign weekly essays, or incorporate separatewriting "units" in an English course. Educational research and common sense suggest. of course,that some choices are better than others, depending on a teacher's circumstances. Whereas traditional (i.e., disciplined) knowledge is the best sourceof curriculum, traditional teaching methods are not necessarily the best sourceof instructional practice. Wigginton, who originated and published theFoxfire books with his students, for example, despaired of the traditionaltools of instruction and developed a method that is clearly not traditional.His students do not work only in classrooms, and they do notproduce the standard sorts of student essays; yet his students learn to writeand to think well. Obviously, teachers need both sets of knowledge, knowledge ofthe disciplines and knowledge of teaching methods. However, without thefor- mer. the latter is useless. ln fact, wedoubt if expertise in teaching method is possible without disciplined knowledge, just as we suspect thatsophisti- cated thinking and reasoning cannot exist without disciplinedknowledge.

Curriculum and individual differences.In the United States educators have placed a great deal of emphasis onaccommodating individual differences within the generality of a common curriculum for all students. There are ideological and practical reasons for thisdevelopment. The traditions of our culture bequeath to all of us a belief in the importanceof liberty and individuality. Appreciation of these values, for example, is at the heart of support for free enterprise ':he United States. At the same time, however, a counter tendency has influenced public educationin the United States. During the early and middle nineteenth century our "common schools" wereviewed as "melting pots" to Americanize immigrants and as agencies to dissolve the distinctions of social class (Katz, 1968; "ryack, 1974). 351 Md hods of Talent Development 343

During the late nineteenth century, educators cameincreasingly to realize that schools did not treat children very muchlike individualsor very much like children,for that matter (Cremin, 1961). Progressive educators (including reformers likeDewey) saw the shortcomings of the public schools and proposed a "chiki-centered" approach toboth curriculum and instruction. Part of this approach entailed observation andaccommodation of individual differences. Over the years. many American educators havemisunderstood what was meant by "child-centered."At one point. Dewey himself impatiently noted that he had seen many "really stupid" responses tohis proposals (Cremin, 1961, p. 259). Dewey did not intendthat curriculum be constructed for each child, or that each child's interestsshould determine a curriculum. Instead, he intended that children'snatural proclivities for learning should be recognized and exploited by teachersin their efforts to tnake disciplinedknowledge accessible to children (Dewey, 1902/ 1956). Although teachers must be sensitive to children's interests,their role as educators is to lead children outof a naive conception of the world into a richer. better-informed, and moresophisticated conception. They cannot fulfill this role if they putchildren in charge of determining the curriculum.

Implications for gifted students.Three implications can be drawn from the preceding discussion. These implications help defineimportant features of the curriculum and of instructional practice for giftedstudents. First, curriculum based on disciplined knowledge is appropriatefor gifted students, as it is for almost all other students. As we notedearlier, because giftedness indicates an exceptional potential for academiclearn- ing, however, the academic curriculum is the curriculum ofchoice for gifted students. Other curricula may be especially appropriate forother types of children. The curriculum for severely handicapped children,for example, usually includes self-help skills. Second, like other children, gifted students need instruction that piques their natural curiosity and provides challengesthat build lxnh knowledge and self-esteem. They do not, any more than other children, need to have their interests consulted in the construction of acurriculum that is tailor-made fbr them. They do not need a "giftedcurriculum." Third, gifted students can typically master the academiccurriculum in several years less time than other students. Instructional practice,therefore, should provide for the acceleration of these children throughthe academic curriculum. Other instructional modifications (certainkinds of teaching strategies, of special courses, of grouping arrangements, ofin- structional materials) can also help gifted students achieve theiracademic potential. 352 344 Methods of Talent Devekposent Curriculum for the Gifted The previous discussion indicates that teachers of the gifted need to know what elements of the academic curriculum gifted children should encounter and when they should encounter them. Readers should remember, in considering the following discussion, that gifted children vary as much as other children. This discussion concerns what is typical, not what is universal. Many gifted student.s will need instructional modifications if the following guidelines are to be implemented responsibly.

Elementaly kvel.Research about gifted students suggests that it is reasonable to expect that many such studentsperhaps (30 percentcould begin high school at the age of 12 if they received suitable instruction in elementary school. Basic skills should be taught up to about age tenat as much as twice the pace of ordinary instruction. By about the age of ten, most gifted children will no longer require instruction in reading per se; and, with guidance from an expert teacher, they can begin to consider serious literature. We recommend that the literature curriculum be established by a committee of teachers in a district. The committee should establish a reading list from which to draw perhaps four of five complete works for a year's classroom work in literature. En- glish instruction should also involve regular writing assignments that move students toward the production of good quality essays of 1,500 to 2,500 words in length by about the age of 12. By about the same age, gifted students should have mastered fractions, dedmals, and measurement concepts. Many gifted students will lw ready to start algebra at this time. Before beginning algebra, however, students should have mastered the critical concepts of proportion and ratio. These concepts are related to fractions and decimals, but they need to be understood conreptually if students are to succeed in algebra. That is, students should understand the meaning of proportion and ratio. They should, for example, be able to appreciate the way in which proportions resemble analogies. A two-year algebra sequence (or a half-year pre-algebra and three-semester algebra sequence) is one way to provide algebra to young gifted and mathematically able students. Foreign languages are not taught well in American schools because foreign language instruction is not considered essential. Nevertheless, we think that, beginning in the elementary school, the curriculum for gifted students should include foreign language instruction. To accomplish this goal, it might be a good idea for teachers of the gifted at the elementary level to be prepared to offer instruction in one foreign language. It is a reasonable instructional goal to prepare many gifted children to enter high school with one year's credit in mathematics (a first course in Methods of Talent Development 345 algebra), in foreign language (one year's credit), and in English (one year's credit). Fulfillment of this goal would, in most cases. require the persistent effort of a dedicated group of teachers, and the assistance of an administrator who understood why it was rea3onable. Secondary level.Many, if not most, gifted children should complete high school (i.e., Grades 9-12) in three years. A numbtr of accelerative options are available to implement this recommendation. A critical educational issue confronts the adults responsible for gifted students during the high-school years. At this time, many able students begin to internalize an academic self-image. Some will be generalists, and they will consider themselves equally interested in mathematical disciplines and in the humanities (literature, history, and the arts). An equal number, however, will want to specialize in one kind of study (e.g., mathematics courses or literature courses). Yet, students of this age (12-16) need io acquire a background in both mathematical disciplines and in the humanities. It is essential to establish a distribution requirement to guide students' selection of courses. As we have suggested elsewhere (liowley et al., 1986), a 2 to 1 ratio between courses in more favored disciplines and less favored disciplines seems to make sense. This provision is important to observe, because it would be possible, in constructing flexible instructional practices for gifted students, to overlook the need to provide program balance. Finally, gifted students in high school should follow programs that expose them to original source.% in history and literature. Part of such exposure should include regular writing assignments about the works that are read. We can find no better preparation for work in competitive colleges and universities. We also believe that a second course in algebra (which includes trigonometry) is the minimum level of course taking in mathematics that should be required of gifted students; nmst gifted students, however, should also take one year of calculus during high school.

TheArt3 The study of art and music does not fare well in v-ir schoolssub- stantially worse, perhaps, than the stu,ly of foreign languages. Neither the arts nor foreign languages is considered veryuse4ill, but knowledge of each is critically important to a deeper understanding of the world. All children should learn about the arts; and gifted students seem particularly interested in the philosophical (aesthetic and metaphysical) implications of the arts (Tidwell, 1980).

The arts and artists in society.The lack of a well-defined curriculum in the arts has been cited as one of the reasons fin the lowly status of the

35 4 346 MethodsofTalent Devehaposetst arts in our schools. However, thereare other reasons that the arts are not supported and do not receive the attention that readingand mathematics receive. The arts are notso useful as reading and mathematics, and the American ethos values most those skills thatare most useful. Artists tend to be aberrant individuals inour society and, in fact, in most Western societies. They live unconventional lives; they articulate views that offend those inpower; and they often produce works that challenge common sense and populartastes. Very few of us would be pleased to see our children grow up to be artistsor musicians, if only because we know that such lives areso often filled with privation and hardship. At the same time, many students, including largenumbers of the gifted. express an abidingconcern with aesthetics. Few of these students will seek to become professional artistsor musicians, but the aesthetic interests of most will persist throughout their lives. Formany of these students, the arts will become asource of continuous intellectual and emotional strength. FAlucators concerned with such studentsshould recognize these facts and assist in the improvement ofarts programs in the schools.

Improvement in school arts programs.More instruction' in the arts, particularly in the history of the arts, should he provided in puolic schools. Many art and music teachers in our schools today have nearly impossible jobs; some see hundreds of students (sometimes more than LOW, They rarely teach the history of art or music in much detail. It is unfOrtunate that the (ally images and sounds most children grow up withare those found on television, billboards, and popular radio stations. Art and music instruction are as intellectually inq)ortant as literature. mathematics, history, and natural science. Art and musiccan help all students attain a quality of' life that is not directly accessible in other disciplines. Those students who are exceptionally talented in the arts also suffer from the fact that the arts are neglected in our schools. Such studentsmust receive their most significant instruction (usually musical instruction) pri- vately. Teachers of the gifted can contribute most to the education of' exceptionally talented arts students by servingas advocates. In this role teachers can support school arrangements that accommodate thesesw- dents' need for private instruction. In most cases, tests and assignmentscan be rescheduled or altered to accommodate the gifted student's practiceor performance schedule. For high-IQ and high-achieving gifted students, giftedprograms can incorporate minicourses on art history and musicor perhapsa series of such coursesto help demonstrate to schools what is needed in aesthetic education. In addition to the gifted, such courses could include other able students. or %).)t) Methods of Talent Development 347

CHARACTERISTICS OF INSTRUCTIONFOR GIFTED STUDENTS The topics of instruction and ofcurriculum are, of course, related. The point of the previous section wasthat the regular curriculum, conceived as disciplined knowledge organized forinstruction, is the curriculum most appropriate for gifted students. Wehave emphasized throughout thedis- cussion in this chapter, however,that the way in which the curriculumis presented to gifted children shoulddiffer from the way it is typically presented in regular classrooms. The most efficient andso far as weknowthe most effective way to accomplish this goal is throughacceleration. At the same time, acceleration falls very Lir short of being a complete responseto the question of how to teach gifted children. We know thatgifted students, more than other students, benefit from certain instructionalformats and techniques. Those who teach gifted students, whetherregular educators or teachers of the gifted, should at least be aware of suchfindings. Moreover, teachers of the gifted, in particular, need to conductlessons that make use of these formats and techniques. Finally, these instructional formatsand techniques have nat, in most cases, been developedespecially for gifted students. "I-hey areprobably beneficial to other students as well. Noneof them is harmful to other students. We know only that giftedstudents benefit more from them than other students. "this finding may be more acharacteristic of gifted students than it is a characteristic of theseinstructional formats and techniques. Instructional Formats Instructional formats come in two basic types,expository and inductive. The formats most commonly usedin schools are expository. Lectures, directed reading, programmed instruction,and individual learning activities arc expository formats appropriatefor gifted students. inductive formats are less seldom used than areexpository formats, perhaps because their use requires more effort from teachers.Tutorials, seminars, independent study, simulations, inquiry trainingsessions, and laboratory and field studies are inductive formats appropriatefor gifted students. Inductive formats for teaching are presumed tolead to more active learning. Expository formats, on the otherhand, sometimes fail to involve students in active learning as much as theymight. Whereas induction is the construction of meaning for oneself, expositionis a reconstruction intended to make meaning accessible to others.Obviously, it is more difficult for teachers to induce learning than to expoundknowledge. Similarly, practicing induction is often more difficultfor students than listening to an exposition. 348 Method, of Taira Dovekpouist Induction is the essential act of learning; exposition is the essential act of teaching. From the viewpoint of the one who learns, almosteverrhiag that is learned is experienced as induction. From the viewpoint ofthe one who teaches, almost everything that is taught is experienced as exposition. In the classroom, the two should go hand-in-hand. Goadteachers see that they do. In the discussion that follows, educators should recallthat both induction and exposition are essential activities of the human mind.The most effective instruction blends both inductionand exposition.

Which format is best?Answering this question depends on a teacher's sense of self',stucl...ms, curriculum, and instructional aims. Most research on teaching suggests that almost any methodselected freely and used vigorously by a koowiedgeable and experiencedteacher will be effective. At the same time, we know that when teachers are given the opportunity to select ways to teach a part of the curriculum, they do notproceed on the basis of whimsy. Nor do they allmake identical choices. In making their choices. they think about such things as the following:

the instructiimal iwtivities that have been nmst successful for tlwm in the past (teacher's instructitmal preferences); the students in their class nowtheir needs and the quality of their classroom interaction with one another and with the teacher (students' characteristics); the nature of what is to be taught (cc. -iculum, instructional oliiectiws of the program); and. fmally, the time, place, and materials available to them (instructional resounes of the pmgram).

Such reflections determine the answer to "Which format isbest?" 'that is, no format is best in all cases, and each can serve a useful purpose.

Inductive formats.Inductive formats are thought to help students "induce" their own learning. Actually, when inductive formats are successful, students are led by their teachers carefully through a series of challenging events, propositions. or paradoxes. Through activities such as these. students. if successful, acquire new concepts. They often do, and well- constructed inductive learning experiences are memorable. both for students and teachers. Inductive formats are especially pertinent to concept development, because in successful inductive activities, students participate activelyin the development of the concept. Good inductive lessons often requirethat students engage in cycles of analysis, synthesis, evaluation,comprehension, and application (Bloom et al., 1956). Hence, inductiveformats seem to meet the need to teach "higher-order thinking" togifted students. 35 7 Methods of Talent Development 349 Whether this need is actually met by inductive formats is difficult to determine. Clearly, all students need practice in such intellectual work, and it is equally clear that much of their time in school is not spent in such activity (Marzano & Hutchins, 1985). Inductive formats are a powerful tool to help teach important concepts to students. Gifted children will benefit greatly from some inductive instruction. However, inductive formats are not an alternative to gifted programs based on acceleration; they are not a temedy for inappropriate placement; and they are not appropriate for teaching many basic skills.

Expository formats.In general, expository formats are more efficient than inductive formats. Expository lessons are also easier for teachers to prepare than inductive lessons. These two facts account for the popularity of expository formats. In fact, part of the appeal of inductive, as opposed to expository, formats centers on one's view of efficiency. The "cult of efficiency" has been viewed as an unhealthy and inhuman influence in American education (Callahan, 1962; Silver & De Young, 1986; Tyack, 1974). Inductive formats incorporate the view that efficiency is not the most important educational value (cf. Taba, 1965). And yet, with gifted students, educators should recognize that considerations about the efficiency of instructionor at least the efficiency of' learningare important. Time is too short for gifted students to learn everything for themselves, even if they might. "Learning by doing" is important, but so is learning by reading and learning by listening. It also makes good sense to teach many things, especially basic knowledge, using expository formats. For example. though students could learn the multiplication facts inductively, massed drill and practice is probably a more efficient and effective way to help students memorize these facts. The same point can be made about many topics in language arts, geography, science, and foreign languages.

Caveat magister (teacher beware).Educators concerned with gifted children should be wary of the claim that inductive formats are the method of choice for gifted students. 'This claim overstates researdi findings that show gifted children to benefit more than other children from this type of instructional format. Though inductive formats provide an opportunity to engage in uncommon instructional activities, they will nurturehigher-order thinking only to the extent that they are used in an overall scheme of curriculum and instruction that makes sense. No research exists that describes the features of such a scheme. We do know, however, that a comprehensive program rarely, if ever, exists for gifted students (Cox et al., 1985). In a challenging academic program at the elementary level. no more ,3)S 350 Methods of Talent Development than about 25 percent of the instructional time should be devoted to inductive formats. In a challenging academic program at the upper secondary level (ages 14-16), some gifted students could perhaps spend as much as half their instructional time in tutorials, seminars, and laboratory work. lf, however, the academic program for gifted elementary and secondary students is less than rigorous, these recommended proportions are too high. On the other hand, teachers who do not incorporate inductive routines into their teachingand short inductive routines can be incorporated even in lectureswill certainly fail to engage students actively in learning.

Instructional Techniques Instructional techniques are the events through whkh teaching actually takes place. The same influences that determine instructional format can also affect the instructional techniques a teacher chooses. These influences can be the teachers' instructional preferences, the students' characteristics, the curriculum and objectives of the program, and the instructional resources of the program. Either an expository or an inductive instructional format can accommodate similar instructional techniques. Inductive formats may incorporate techniques that call for explanations (exposition) by the teacher, and expository formats may incorporate techniques that call fbr open-ended (inductive) responses from students. However, for any instructional technique to be efkctive in a program for gifted students certain conditions must be in place:

careful selection of students, program dements flexibly eninigh to address individual needs, a tnatch between curriculum level and students' achievementlevel, a curriculum relevant to Sitstirnis. longtermeducatiimal goals. and a teacher who serves as an intellei Ina! nuidel.

Teacher as intellectual model.Tlw first four conditions were discussed in previous sections of this .apter. The last is a critical concern with respect to instructional techniques. Expertise in disciplined knowledge is the basic requirement of teachers of the gifted (Gallagher, 1975). Expert teachers are the only ones who can implement successful academic programs for the gifted or develop existing programs further. Of course, expertise in an academic discipline is necessary in order to manage any academic program well or to engage in sensible program and curriculum development. Teachers of the gifted should serve as intellectual models to their students. They should, for example, write (Koch, 1973), paint (Chetelat. 1981), or engage in continuing studies of their own (Brown, 1967). What-

:3 5 Mrthods of Talent Development 351 ever else they do,they should also continue to readwidely. Cohn (1984) reports that suchteachers influence not only theachievement of their students, but their intellectualvalues (appreciation of intellectual work) as well. Perhaps one of the strengths ofthese teachers is that they have a well- informed sense of teachingmethodinstructional preferences that show a high degree of self-awareness, adiscipline-relevant understanding of student characteristics, a knowIrdgeof subject and flexibility in determining objectives, and a clear understandingof essential resources. Such an informed sense of teaching methodhelpstheseteachers establish successful instructional routines.

Techniques for concept development. The doi4unent of concepts in students, in all of us. really, is asubtle process of evolution. Though we refer to concepts by words (e.g.,"literary genre." "apartheid," "input-out- put economics," "entropy") wedo not develop concepts by memorizing definitions. Original thinkers, for example. canlegitimately understand each of the preceding concepts in verydifferent ways. That is, thinkers induce their own concepts. Readers mustrealize that, in some sense, all human beings are original thinkers. The concepts towhich we refer here, however, are primarily academic concepts. Students' concepts are often quite naive.This observation applies even to successfulstudents (Mestre & Gerace, 1986).According to Mestre and Gerace, students' miscimceptions arecommonplace. That is, students are prone to the same sortsof misconceptions. When teachers seek to correct thesemisconceptions, they are really engaging in the business of conceptdevelopment. Note that concept development in this view israther like the behaviorist notion of"successive approximation." It respects students asthinking beings, and it is based on confidence in students' ability to refinethe concepts they possess. Successful concept development canemploy both expository and inductive techniques. An expositorylesson:

supplies a definition. supplies exampks, and contrasts examples with closenonexamples.

Note that these activities can becarried out with no active student participation at all. However, a basicallyexpository format (e.g., a lecture format that also uses expository techniques)could be amplified to include an inductive routine in which students proposeexamples and nonexamples. As part of this activity a class discussionmight result in the generation of related concepts to account for closenonexamples.

3 t;) 352 Methods of Talent Development Inductive techniques for concept developmentlead students to abstract rules that define a concept.Instead of being told what the rules are. students:

differentiate the qualities of objects and events, abstract shared qualities, and categorize objects and events on the basis of theirobservations.

Teaching about concepts with inductivetechniques relies on good examples of the concept. In this case "examples" are notusually verbal descriptions of objects or events. The examplesused in inductive techniques are typically real-lifesituations in the environment. Suchman's(1975) inquiry training sessions, fbr example, use aninductive format in which students confront scientific paradoxes (real-lifeexamples) that seem to defy common sense. Studentsabstract the concepts that explain theparadox through a carefully structured series ofobservations and questions, (inductive techniques). Whenever possible,inductive techniques bring the "real" world into the classroom. Inductive lessons arethought by some educators to be a good antidote fc,rschoolwork that relies too heavily on printed materials and on rote learning. Inductive techniques include a good many"hands-on" activities. Hands-on activities involve students with concreteexperiences and manipulative skills. Students acquire "operational" concepts(that is, nonverbal concepts) by observing what happenswhen they manipulate the environment. This characteristic ofinductive techniques suggests that concrete experience informs even verbal conceptdevelopment. Concrete experiences and abstract perceptionsmake up a completely understood concept. In a similar way, induction andexposition make up a whole mental act. (Note that we saidpreviously that students "explain" paradoxesin Suchman's inquiry training sessions.) Theend product of induction is an explanationa specifically formulated statementthat can be communicated verbally to another person. Finally, teachers should be aware of theempirical research on question asking. Good questioning skills areessential techniques for either inductive or expository lessons. The kind ofquestioning that nurtures students as thinkers is very different, forexample, from questioning used to control student misbehavior. In a way, the tc-iiques of questioning that help students think are teachers' mostimportant tools (cf. Adler, 1982). Here are some suggestions for teachers,based On the research about questioning:

Do not convey the impression that classioomquestions are a kind of test in which answers %AI be judged quickly for correctne-ss. Questions that help dispel misconceptions or build new concepts are nem- Methods of Talent Development 353 tentative- sarily ambiguous to students.Accommodatestudents' uncertainty or ness. Give individual students ampletime to respond to questions. seamds, you may wish to rephrasethe question. If you get no reply in fifteen dealing with As much as possible, help thestudent tolerate the uneasiness of necessarily ambiguous questions. raised by the If a student's response is notsatisfactory, address the ideas student in the response, andinvite the student to provide arevised or extended answer. Do not simplyjudge the correctness of thestudent's answer. You can follow up anincomplete response or a responsethat indicates a misconception with a "probe" tohelp the student considercontradictory or misunderstood information. question to several Give the class ample time toresjxmd to a question. Pose the students before you help them toevaluate the answers.

all else, serve to stimulatestudents' thinking. Questions should, above (cf. They should never beused as a techniqueof classroom management questions and judgingthe cor- Gtxxi & Brophy, 1987).A aking rapid-fire control over children,but it is a rectness ofresults may appear to exert counterproductive kind nfcontrol (Anyon, 1980;Giroux, 1983). This mis- that serve work- use ofquestioning is a commonpractice in some schools ing-class students (Wilcox& Moriarity, 1977). Sometimes, of course, ateacher may want toknow if students have questioning helps the understood a point in a lecture orin a reading. Such students' needs. It adjust lesson plans sothat instruction matches t acher used to assess students' should not be overused,and it should nevrr be classroom performance.

Techniques for academic shindevelopment. We impliedabove that the development of skills isrelated to the developmentof concepts. We also by teachers in either noted that conceptdevelopment could be approached mode or in an expositorymode. So it is with skilldevelopment. an inductive development, and The acquisition of newskills leads inevitably to concept new skills mayalso be taught in inductive orexpository modes. There is one importantdifference. Skills are habitualbehaviorsfor example, the skill ofproofing text, or the skillof solving trigonometric equations, or the skill ofdeclining Latin nouns. Skillsbecome habitual only with a certain amountof practice. The acquisitionof a new skill probably depends as much on practice as onanything else (providing, of course,that the new students are conceptually ordevelopmentally prepared to learn skill.) The acquisition of newskills requires exposure toand analysis of skill models. Students need to patterntheir new behaviors aftergood models. A teacher's role is to useprofessional judgment inselecting skill models that 3U2 354 Methods of Talent Development are appropriate for students. A good exampleof this process can be found in Kenneth Koch's (1970) introduction to Wishes, Lies, and Dreams. Koch encourages teachers to consider models thatseldom find their way into the classroom. Koch's work is of the type Stanley (1976a) calls "relevant academic enrichment." The kind of judgment exemplified by Koch, however, needs to be applied to curriculum development across all subjectsand instructional levels. Educators concerned with gifted students also need to consider the issues of pace and comprehensive educational planning when they design units that are intended to teach new skills. Before students can practice the skills we want them to learn, they need to understand the features that make the selected models gcxxl. Again, both expository and inductive techniques can help students reach such an understanding. In geometry, for example, the features of a gcxxl proof can be explained in a lecture and demonstration. On the other hand, students might be asked to compare several proofs, and in the process the teacher could pose questions that lead students to abstract for themselves the features of a good proof. Because students usually practice skills independently, teachers are of-ten concerned to monitor the quality of studems' practice work. Very often, teachers assign grades for practice work. This approach to students' practice is not defensible, however (Good & Brophy, 1987). There are better techniques for monitoring the quality of students' practice work. Among these techniques are the following:

giving ungraded practice quii.a.s to help students evaluate their own progress, selixting appropriate practice materials that t hallenge students and lhive the potential to engage their interest. using task cards to differentiate assignments (Good 8.7 Bmpliv. using peer tutors, using cooperative karmng technipies (Slavin. 1986), kiting students complete wiitten assignments using word processing. using thstructional software, and establishing classroom routines that give students ampk opportunim s. to ask questions of the teacher.

SUMMARY

This chapter provided an overview of methods for educators concerned with gifted students. The discussion was guided by empirical research about teaching these students. This rewarch concerned alternative meth- Methods of Talent Development 355 ods of identification, methods to ensure appropriate placement and program planning, methods of acceleration and enrichment for gifted students, and methods of curriculum and instructional design thought to be appropriate for gifted students. We acknowledge, however, that this approach has its shortcomings. Therefore, at the end of this chapter and of this hook, a note about the place of' methods in the real world of the classroom seems to be in order. Despite our insistence on heeding the findings of research, we believe that teaching is more art than science. Nathaniel Gage (1978) has summarized the "scientific basis" of' the art of teaching. If you read his work, you will perhaps agree that the best that can he said about the scientific basis of the art of teaching is that there is one. Whether it inf'orms the practice of' teaching is another matter. And the ordinary practice of teaching is a bit different from the art of teaching. The art of teaching is the adept practice of teaching. Excellent teach- erslike gifted childrenare comparatively rare. Clearly we need to know more about them than we do at present; we would all like to be excellent teachers. Most of us who work hard, however, do get the chance to be considered good teachers by our students. Very few of us, we suspect, are capable of' being consistently excellent. The good teachers we have known do many of the things cited in this chapterthey have a grasp of disciplined learning, they make good plans, they ask questions well, and they make students feel cared for. They achieve these ends, however, in many different ways. Some favor inductive methods, some stick to expository methods, and still others improvise an effective combination. All of them put a great deal of effort into their work. Teaching is the kind of joh that takes all your skill if you wish to bc good at it, no matter how talented you might be. No matter how talented you might he, there is always more to do, and there is always some portion of Our routine to improve. In addition, the conditions of a teacher's work usually change over the course of a career; teachers often encounter different neighborhoods. different students, and different colleagues in the course of a career. They may accept different teaching assignments at different levels. Perhaps a teacher's hest resource in all of this flux is a sense of the alternatives: how to change a particular feature of a lesson, how to offer alternate explanations to students who seem to have missed the point, how to adapt a new technology to good instructional purpose, or how to switch from an expository to an inductive approach in the middle of a lesson. Such resourcefulness looks a lot like creativityfluency, flexibility, originality, and elaboration. Take heart! Research on creativity says it is easy to learn! In any casekidding asidea holistic sense of method is preferable

3 f 4 356 Methods of Talent Development to an atomisfic, textbook sense of method. A holistic sense of method t an be acquired by all teachers to some degree over time. It is also thesort of thing that can only be acquired inductively;so we hope that you will accept our expository approach (a textbook) as an invitation to read, act, and reflect further on the nature and needs of students, giftedor not, and on the educational arrangements that best accommodate theirnurture. References

Ator.soN, P. (1967). Relation of group activity, to creativity in science. In J. Kagan (Ed.). CTrailVal and learnmg (pp. 191- 202). Boston: Beacon Press. ADAMS. J. E.. & Ross, C. 0. (1932). Is skipping grades a satisfactory method of acceleration; American School Board Journal. 85, 24-25. ADLER, M. (1982). The Paidra p oposal: An educational manifesto. New York: Macmillan. ALBERT, R. S., & RuNco. M. A. (1986). The achievement of eminence: A model based on a longitudthal study of exceptionally gifted boys and their families. In R. J. Sternberg & j. E. Davidson (Eds.). Conceptionsof giftedness (pp. 332-357).Cambridge: Cambridge University Press. ALBRIGHT, H. (1967). Acting: The creatwe proms. Belmont. CA: Dickenson. ALEXANDER. P. A. (1985). Gifted and nongifted students' perceptimis of intdligence. Gifted Child Quarterly. 29(3), 137-142. ALEXANDER, P. A., & SKINNER, M. (1980). The effects of early entrance OD subsequent social and academic development: A f011ow-up study. Journal for the Education of the Gifted. 3(3), 147-150. ALEXANDER, R. (1981). An historical perspective on the gifted and the talented in art. Stutirei ui Art Education, 22(2), 38-48. ALc;OZZINE, B.. VSSELDSBE, J., & SHINN, M. (1982). Identifying children with learning disabilities: When is a discrepancy severe? Journal of School Pricholop, 20(4). 299-305. ALLEORT. C., VERNON. P., & (1951). Study of tydur.s: Manual of &realms. Boston: Houghton Mifflin. ALTER, J., DENMAN. D., & BARRON, F. (1972). Dancers write of themselves and dance education. In F. Barron (Ed.). Antics in the malting (pp. 86-112). New York: Seminar Press. ArviNo, J., McDONNEI., R. C.. & Ru:nxicr, S. (1941). National survey of idemifkation prac tic es in gifted and talented education. Exceptional Children, 48(2), 124-132. Al-VINO, J., & WEILER, J. (1979). How standardized testing fails to identify the gifted and what teachers can do about it. Phi Delta ltiappan. (51(2), 106-109. Mils, 1.. B. (1968). Learning disabilities: The developmental point of view. In H. R. My- klebust (Ed.). Progrem in karnnig disabihties, (Vol. I. pp. 39-74). New York: Grune & Stratton. ANASTASI. A.. & SCHAEFER, C. E. (1971). Note on the concepts of creativity and intdligence. Journal of Creative Behavior, 5(2). 113-116. ANDREWS, F. M. (1975). Social and psychological litmus which influent e thereatice process, In IA. Taylor & J. W. Getzels (Eds.). Perspedives in arativity (pp. 117-145). Chicago:

ANot.m. S (1979). Personality, family, educational, and criminological characteristic s of bright delinquents. Psychological Reporb. 44, 727-734. ANSRACHER. H. L.. & ANSBACHER, R. R. (1956E The individual psycholop of Alfred Adler New York: Basic Books.

3 f ; 357 358 References

ANTON, J. (1980). Social class and the hidden curriculum of work.Journal of Education. 162( I ). 67-92. ANYON, j, (1980/1987). Social dass and the hidden curriculumof work. In E. Stevens & G. H. Woods (Eds.). fugue, ideology, and education: An introductwn to thesocial foundations of education (pp. 210-226). New York: Random House. ARENDS,1, & KIRKPATRICK. J. (Eds.). (1987). Budding an excellence.Washington, DC: Council for Educational Development and Research. ARENDT, H. (1981). The We of the mind. New York: Harcourt BraceJoyanovich. ARNHEIM, R. (1966), Growth. In E. W. Eisner & D. W. Ecker (Eds.),Readings in art eduratwn (pp. 85-96). Waltham, MA: Blaisdell. ARNHEM, R. (1972). Toward a psywhology of art: Collected essays.Berkeley. CA: University of California Press. Aitouovv, F. (1965). Music and young daldren. New York: Holt,Rinehart & Winston. Mo. K. (1984). What is the source of art? Journal of Education,166(2), 188-195, AsriN. H. S. (1974). Sex differences in mathematical and scientificprecocity. In J. C. Stanley, D. P. Keating, & L. II. Fox (Eds.), Mothematicol talent: DUcimery,description, and development (pp. 70-86). Baltimore, MD: Johns HopkinsUniverity Press. ACNI IN. A. B.. & DRAPER, D. C. (1984 Peerrelationships of the academically gifted: A review. Gifted Child Quarterly, 23(31 129-133, BAulthIAN.J. C.. (19701. The impaa of family baci groundand intelligence on tenth.grode boys. Youth in transition (Vol. 2). Ann Arbor, MI: Braun-Brumfield. BACKMAN, M. E. (1972). Patterns of mental Ethnic. socioecononM., and sex differ,. em es. Amennin Eduratwnal ResearchJournal, 9(1), 112. BAER. D., & BUSHELL, D. (1981). The future of behavior analysisin the sc hools? Cmisidet its recent past, and then ask a difkrent question,School Psychology Review, 10(2), 259-271), BAIRD, L. I.. (1985). Do grades and tests predict adultaccomplishment? Reseto h in Higher Education, 2;(I), 3-85. BALDWIN, A. Y. (1978). The Baldwin identification matrix. In A. Y.Baldwin, G. H. Gear, k J. Lucito (Eds.), Educational planning for the gifted: Overcomingcultural, geographic. and 30aBeCtMeMiC barriers. Reston, VA: Council for Exceptional Children. BAMBERGER, J. (1986). Cognitive issues in the development of musicallygifted children. In R. J. Sternberg & J. E. Davidson (FAh.), Conceptions of giftedness (pp. 388-413)Cambridge: Cambridge University Press. BARBE, W., & MILONE, M. (1985). Reading and writing. In R. Swassing(Ed.), Teaching gifted children and adolescents (pp. 276-313). Columbus. 011: Merrill. BARKER, R., & GUMP. P. (1964). Big school, small school. Stanford, CA:Stanford University Press. BARNETTE, L. A.. & FISCELLA, J. (1985). A child by any other name. .. Acomparison of the playfulness of gifted and nongifted children. Gifted Child Quarterly, 29(2), 61-66. BARRON,F. (1963). Creativity and psychological health. Princeton. NJ: VanNostrand. BARRON, F. (1969). Creative person and creative proms. New York: Holt,Rinehart & Wis.ston. BARRON, F.. & DENMAN, D. (1972).Students in the theater arts. In F. Barron (Ed.), Artists in the making (pp. 75-43). New York: Seminar Press. BARRON,F., KRAUS, j.. & Com, I. (1972), Six highly rewarded studentsin interview. In F. Barron (Ed.). Artists in the making (pp. 19-32). New York: Seminar Press. BAR-TAI., D. (1978). Attributional analysis of achievement-relatedbehavior. Review of Educa- tional Research, 48(2). 259-271. BARTHE,C. L. (1980, May). Program for academically talented students. Paperpresented at the Annual Meeting of the International Reading Association St. Louis,MO. BARYLICK, M. (1983). Both artist and instrument: An approach todame education. Daedelu.s, 112(3), 113-127, BAR1.17N, J. (1965). Mimic in American life Bloomington. IN: IndianaUniversity Press. Bross, B. ( 1981). Slogdill's handbook of leadership.New York: Free Press. BATEMAN,B. D. (1965). An educator's view of a diagnostic approach to learningdisorders. In J. Hellmuth (Ed.? Learning disorders (Vol. I). Seattle, WA: Special ChildPublications. BavEcas.A. (1984). Leadership: Man and function. In W. E. Rosenbach & R. L.Taylor (Eds.), Cantrinporary issues in leadership (pp. 117-123). Boulder, CO: Westview Press. `0" References 359

BavuLY. N. (1970). Development of mental abilities. In P. Mussen (Ed.),Carinuhaets monad of duld psychology(pp. 1163-1210). New York: Wiley. BELE, E. 'I'. (1965). Menofmaihentaties. New York: Simon & Schuster. BENBOW,C. P. (1983). Adolescence of the mathematically precocious: Afive-year longitudinal study. In C. P. Benbow & 3. C. Stanley (Eds.),Academicprecocity:Aspects of di development (pp. 9-37). Baltimore, MD: Johns Hopkins University Press. BENsow, C. P.. PuouNs, S.. & STANLEY, J. C. (1983). Mathematics taught at afast pace: A longitudinal evaluation of SMPY's first class. In C. P. Benhow & J. C.Stanley(Eds.). Academic preroca":Aspeco of itsdevelopment(pp. 51-78). Baltimore. MD: Johns Hopkins University Press. BEnsow.C.. &STANLEY, J. (1980).Sex differences in mathematical ability: Factor artilactl:, Science. 21(1(12).1262-1264. BENBOw, C., & STANLEY, J.(1981). The devil's advocate: Sex difkrences in mathematical reasoning ability.Journal for the Education of the 64ted, 4(3),169-176. 239-243. BENBow, C.. & STANLEY, J. (1982). Intellectually talented boys andgirls: Educational profiles. Gifted Chdd Quarterly, 26(2),82-88. BENNETT, W. J. (1986). Whatworks: Research about tearhoig and learning. Pueblo.CO: Office of Educational Research and Improvement. BERG. 1. (1972).Rich man's qualifications for poor man's jobs. In R. Lejune Clamand conflict in American society(pp. 186-196). Chicago: Markham. BERusccaton. W. (1968).Mathematics: The art of reason. Boston: D.C. Heath. BERNMIN, B.(1961). Sodal sanctum. language, andkarning. Educational Bevan h, 7(3). 163-176. BI.A1,D. (1980). Art andsport. Journal of Aesthetic Educauon. 14(2).69-80. BIRCH. J., TIOSALL, W.3., PEABODY, K., & STERRETT. B. (1966), Schooladnevement and effect of typesizean reading invisually handicapped children(Cooperative Research Proicet No. 1766). Pittsburgh. PA: University of Pittsburgh. BIRNBAUM. M. (n.d.). Educational problems of rural education for the gifted. Inhim for urban/rural giftedltalented(pp. 67-69). Ventura, CA: National/State Leadership .fraM- ing Institute on the Gifted and the 'I aimed and the Oflice of theVentura County. Superintendent of Schools. BLAcHLY, P. 11., DISHER. W.. & BODUNER. G. (1968,December). Suicide by physicians.Bulletin of Suicidology, I-18. Bum% P.. &DUNCAN, 0.(1967).The Americanoccupational umlaute. New York: Wiley. BEAuctERcs, M. S. (1978). Overcoming the sexist harriers to giftedwomen's achievement.In Advantage: Disadvantaged gifted (pp. 7-38).Ventura. CA: National/State Leadership 'training Institute on the Gifted and the Talented and the Officeof the Ventura tpinny Superintendent of Schools. Bun.AND, R., &MICHAEL, W. (1984). Acomparison of the relative validity of a measure of Piagetian cognitive development and a set of conventional paignostic measuresin the prediction of the future success of ninth- and tenth-grade students inalgebra. Educa- tional and Psychological Measurement. 44(4), 925-943. Bums,. B. (Ed.). (1956).Taxinumn of educational objertwes. Illoulhoolt I: Cognitive domain.New York: David McKay. BLOOM, B. (1964).Stability and changein human rharactertinft New York: Wiley. Bloom, B. (1971). Learning for mastery. In B. Bloom, J. Hastings, & U.Madaus.Handbook on formativeand sumnlatiefevaluation of student learning(pp. 43-57). New York: McGraw.

Burma, B. (1976). Humancharacteristics and school learning.New York: McGraw-Hill. Burma. B. (1977). Affective outcomes of school learning. PhiDelta Kappan.59(3), 193-198. &trim B. (1982). The role of gifts and markers in the developmentof talent.Exceptional Children. 48(6), 510-522. Bloom, B. (1985).Ikveloping talentin young proper. New York: Ballantine Books. Buoom, B., &SOSMAK,L (1981), Talent development vs. schooling.Educational Leadership,79, 86-94. BOEHM. L.(1962). The development of conscience: A comparison ot American childrenof different mental and socioeconomic levels.Child Development 77, 575-590,

3 f 360 References

Bow+, D. (1980). X-linkage of spatial ability: A criticalrevkw. Chdd Demlopment. 51, 625-635. Buss. R.. & MerAnsum, M. (1981). The destructive impact of apositive team-building intervention. Group & Organizatianal Studies. 6(1). 45-56. Bowt:N, J. (1972). A history of Western education, (Vol. 1). NewYork: St. Martin's Press, BowEus, J. (1969). Interactive effects of creativity and IQ onninth-grade achievement journal of Educational Measurement, 6(3), 173-177. Bowtxs, S.. & Gums. H. (1976). Schoohng in rapnahd Anterim.New York: Basic Books. Many, F. (1973). Profile of a prodigy: The 10, and games ofBobby Fischer. New York: D?vid McKay. BicAt..s, J. (1971). Early admission: Opinion versus evidence. TheElrrnrraaiy Si loud Journal, 72(1), 35-46. BRANDWLIN. P. (1955). The gifted student as future scientist: The highschool student and ho commi1 . ment to science. New York: Harcourt BraceJovanovich. BRAUNSCHWEIG, P. (1981). Training the young dancer. Design for Arts inEducation. 87(1). 70- 74. BREKKE, B.. JOHNSON, WILLIAMS, J. D., & Mtuonson, E. (1976). Conservatim of weightwith the gifted. Journal of Genetic Psychology, 129, 179-184, BRICKLIN. B.. & BRICKLIN, P. M. (19(i7). Bng1U hild-poorgrades: The pswhology undenulueve- merit New York: Ddacorte Press. Butx.uwAv, W., & WEIN51OCK, 11. (1958). Men of IntRUC Thrtt tifFW, and ac hievements. New York: Simon & Schuster. BRONFENISKINNLR. I.T k CROMER, 11. C. (1982). Work and familythtough time and spate.lii S. B. Kamerman &C. D. Hayes (Eds.). Families that work:Children m a changing world (pp. 39-83). Washington, DC: National Academy PI eSs. BROUDY. Ii. (1958). A realistic philosophy of musk. In N. Henry(Ed.). Basie roneepu in maw education (pp. 62-87). The Fifty-Seventh Yearbook of theNational Society for the Study of Education. Part I. Chicago: University of ChicagoPress. BROWN, A. I.. (1978). Knowing when, where, and how to remember:A problem of metarogni- non. hi R. Glaser (Ed.), Advances in instructwiud phahigi. (Vol1). Hillsdale, NJ: BROWN, J. D. (1967). The development of creativeteacher-scholars. J. Kagan (Ed ), Creativity and karrung (pp.164-180), Boston: Beacon Press. Buticn, C. (1975). Children with intellectual superiority. In J.Gallagher (Ed.). The applwatwn of child development research to exceptionat children. Reston,VA: Council for Exteptional Children. BattnErt, J. S. (1960). The proms of education. New York: VintageBooks. BRYAN, T. IL, & BRYAN, H. J. (1978). Understanding learningdisabilities (2nd ed.). Sherman Oaks, CA: Alfred. BCCHMANN. M. (1987). Repotting and using research. In J.1. Gcxxllad (Ed.), The ecology or school renewal (pp. 170-191). Chicago: National Society for the Studyof Education. BuENNING, M., & Tou.u,sos. N. (1987). The cultural gap hypothesis as anexplanation few the at hievement panerns ofMexican-American students. Porlwlogy in the Schools, 24, 264- 272. Bell.. B. L. (1985). Eminence and precocity: An examinationof the jiistifkatnn»I education for the gifted and talented. Teachers College Record, 87(1), 1-19. Burt, K. S. (1987). The learnmg disabled gifted child: Educationalprwdures in rum/settings. Papet ptesemed at the Seventh Annual Conference of the American (oncil onRural Special Education, Asheville, NC. Beaus. B., JENstiv, D., & TLRMAN. 1.. (1930). Thf pTIMWe ofyouth: Genetic studies of Arita% (Vol. 3). Stanford. CA: Stanford University Press. BuRris. j. M. (1978). Leadership. New York: Harper & Row. Beal, C. (1966). The genetic determination of differences inintelligent e: A study of memo- rygotit twins reared together and apart. British Journal ofStatistical Psychology, 57. 137 153. Boil, C. (1975). The gifted child. New York: Wiley. BIA-VEREIELD, F. (1986. August 3). Why Asialls ategoing to the head of the class. .Ven. loth Timet, pp. 18- 23. firers. K. F. (1973). Thr education of the West. New York:MtGraw-Ifill. Refeertzres 361

CALLAHAN, & RENZULLI, J. (1981). The effectiveness of a creativity training programin the language arts. In W. Barbe & J. Renzulli (F.ds.), Psychologyand education of the gifted (3rd ed., pp. 394-400). New York: Irvington. CALLAHAN, R. E. (1962). Educatthn and the cidt of efficiency. Chicago:University of Chicago Press. CALLAWAY, W. R. (1970). Modes of biological adaptation and theirrole in intellectual development. In The Galion InsWute Monograph Series,No. 1, 1-34. CAMPBELL. J. (1971). The ponabk Jung. New York: Penguin Books. CARIHER, J. G. (1983). Masking the sodal in educational knowledge:The case of learning disability theory. American Journal of Sociology, 88(5), 948-974. CARROLL.J. B. (1963). Research on teaching foreign languages. In N. L. Gage(Ed.). Handbooh of research on teaching (pp. 1060-1100). Chicago: Rand McNally, CARROLL. J. B. (1976), Psychometric tests as cognitive tasks: A new structureof intellect. In L. B. Resnick (Ed.), The nahar of inklligence (pp. 27-56).Hillsdale, NJ: Erlbaum. CAittrit, & KONTOS, S. (1982). An application of cognitive-developmentaltheory to the identification of gifted children. Roeper &WU', 5(2). 17-20. CAR-TER. K., & ()ammo, J. (1982). Acquisition of formal operationsby irnellectually gifted children. Gifted Child Quarter/y, 26(3), 110-115. CARTER, R. S. (1952). How invalid are marks assigned by leathers?Journal of Educational Psychology, 43, 218-228. CARTWRIGHT, G. P., CARTWRIGHT C. A., & WARD, M. E. (1981). Educatingspecial framers. Belmont, CA: Wadsworth. CAssEstLy, P. L (1979), Helping able young women take math and scienceseriously in school. In N. (olangelo & R. T. Zaffrann (Eds.), New voices in rounsthng the gifted pp.346-369). Dubuque. IA: Kendall/Hum, CAscit:LIONE, I.. V. (1984, June 6). The education of the gifted andtalented: Nurturing a natural resource. Education Week, p. 40. CAcoN, H. R. (1985). Visual impairments. In W. H. Berdine & A. Fl. Blackhurst(Eds.), An introduaion to special educahon (2nd ed., pp. 235-282). Boston: Little. Brown. CA-ion. Ii. R., & RANKIN, E. (1980). Variability in age and experience amongblind students using basal reading materials. Journal of Viswd Impairments and Bhndneu, 74,147-149. CA-rut-L., R. B. (1950). Culture fair intelligence Srak I. Champaign. IL: Institute for Pet- sonality and Ability Testing. CArrELL, R. B. (1971). Akfines: Their structure, growth, and action.Boston: Houghton Mifflin. CATTELL. R. B., & CArrELL. A. K. S. (1960). Culture fair inttlhgence test: Scale 2.Champaign, IL: Institute for Personality and Ability Testing, CArrELL, R. B., & CATIELL. A. K. S. (1963). Culture fair Intelligente test:Scale 3. Champaign, IL: Institute for Personality and Ability Testing. I:HALL...J. (1967). Learning to react. The great debate. New York:McCraw-Hill. CHIARI, J. (1960). Realism and imagination. London: Barrie and Rocklin. CHANSKY, N. (1964). A note of the grade point average in research.Educational and Psychologr. cal Measurement. 24, 95-99. CHETLLAT, F. (1981). Visual arts education for the gifted elementary level artstudent. Gifted Chad Quarterly, 21(4). 154-158. CLARizto, H. F., & MEHRENS, W. A. (1985). Psychometric limitations ofGuilford's structureof-intellect model for identification and programming of the gifted. Gifted Child Quar- irr4, 29(3), 113-120. CLARtzto, H. F., & PHILLies, S. E. (1988). Sex bias in the diagnosis oflearning disabled students. Psychology in the Schools, 23(1), 44-52. CLARK, B. (1979). Growing up Ord. Columbus: OH: Merrill. CLARK. B. (1983). Growing up gifted (2nd ed.). Columbus, OH: Merrill. CLARK. B. (1986). Ordinsizing karning: The integrative education model In thedawaam. Cohimbus, OH: Merrill. CLARK, B. (1988). Growing up gifted (3rd ed.). Columbus, OH: Merrill. CLARK. G.. & ZIMMERMAN, F.. (1983). Identifying artistically talentedstudents. tirlwol Arts, 83(3), 26-31. CLARK, G.. & ZIMMERMAN. E. (1984a). Edurahng artistically kilenkd students.Syracuse, NY: Syracuse University Press. 362 Rilieronees

CLANS, a. 8t ZIMMERMAN, E. (1984b). TOViani anew concepfion of talent in the visual arts. Ruler &vim, 6(4), 214-216 CLANks.. E. L. (191611968). Ainerkan men of letters: Thnr nature and nurtum NewYork: AMS Press. Cuss Esrrs, S. D. (1966). MOansal bnzin dysfunction in children: Tnininologyand identification, Phase one of a three phase prukci (NINDS Monograph No. 3, U. S. PublicHealth Service Publication No. 1415). Washington, DC: U.S. Government Printing Office. COHN. S. J. (1984). Project F. U. T. U. IL E.: A progress report after three prototype yearsof servmg intellectually gifted students. Journal forthe Education of tlw Gifted, 7(2), 103- 119. Cot.Asnal.o, N., Ik PELEGEP., L. R. (1979). Academic self-concept of gifted highschool students. In N. Colangelo & R. T. Zaffrar a (Eds.), New voices in counseling the gifted (pp. 188-193). Dubuque, IA: Kendall/Hunt. CotistAN, J. (1961). The adokscent society. New York: Free Press of Glencoe. COLEMAN, J., CAMPBELL, E.. HOBSON, C.. MC.PARTLANO, .1.. MOOD, A.. WEINFELD. F.. &YORK. R. (1966). Equably of educational opportunity report. Washingtk n, DC: U.S.Government Printing Office. COMMITTEE ma EcoNe,441c DEVELOPMENT.(1985). Investiv in our children: Bwiness and the public schools. New Vark: Committee for Economic DevelopriP.m. Corm', F. 0. (1961). The Ant/tic-an high school and the talented student. Ann Arbor. MI: Univer- sity of Michigan Press. Cogitate., D. (;. (1984). Families of gifted chihiren. Ann Arbor. MI: UM1 Research Press. (OUNLIL OF Si Alt DIRE.CTORS OF GIFTED PROGRAMS. (1985). The state of thr states'gifted and talented education. Augusta, ME: Maine Department of Education. Cottsrrs. G. (1930). The Amerwan road to culture; A social interpretation of education in M.United States. New Yi,rk: Jnhn Day. (Facsimile reprint edition by Arno Press-New York Titnes. 1971) Cottsrrs, G. (1932). Dare the sclumi build a new social order? New York: John Day. Cox. C. (1926). The early mental traits of three hundred geniuses: Genetic.studwsf genius (N 'ill. 2). Stanford. CA: Stanford University Pre-.. Cox. I.. & DANILI.. N, (1983). Specialized schools for high abilitystudents. (.://'1T. 28. 2-9. Om J., DANIxt.. N.. & BOSTON, B. (1985). Educating able /earners;Programs and promising prat en. Austin. 'EX: University of Texas Press. CRIMIN. 1.. (1961). The tmnsforrnatwn of the school. New York:Vintage Books. Carson. L. (1980). Amman education: The national experience. NewYork: Ilarpei & Row. Clams. D. P. (1985). Phy5kal and health-related disabilities. InW. H. Berdine & A. E. Black- burst (Eds.). An introdurtum to spenal education (2nd ed., pp. 283-332).Boston: little. Brown. (:tliAst. L. (19)42). Persistence of ..be inevitable: The teacher-centeredclassroom. Education and Urban Society. 15(1). 26-41. Ci Nummis. S. Bainos, IL. & Wary. J. (1977). Preachers versus teachers:Local-cosmopolitan conflict over textbook censorship in an ApFa1achian community. Rural Socwlogy,42(1), 7-21. Cilit IN. M., AvNia, A.. & SMITH, L. (1983). The Pimsleur Battery as apiedktor of student performance. The Modern Language Journal, 67(1). 33-40. Dm:Ly. J. S.. & MADAUS, C. F. (1971). An analysis of two hypotheses concerningthe relationship between creativity and intelligence. Journal of Educational Research, 64(5). 213-216. DAlawits. (1859/1958). The origin of +me,. New York: Mentor Books. (Original work published 1859) DAUR1O. S. P. (1979). Educational enrichment versus acceleration: A review ofthe literature. In W. George. S. Coht & j. Stanley (Eds.). Educating the gifted: Acceleratton and enrichment (pp. 13-63). Baltimore. MD: Johns Hopkins UniversityPress. DAVIDSON. H. P. (1931). An experimental study of bright, average, and dullchildren at the four-year mental level. Genetic Psychology Monograph. 9,119-289. DAvnisor. ff.. & BALDUCLI, D. (19r-6), Class and sex differences in verbal facilityof bright children. Journal of Educational Psychology, 47(8). 476-480. DAVIS, c.. & Ram S. (1982). Group inventory for finding interests (GIFF1) 1 and 11:Instrn- mem for identifying creative potential in the juniorand senior high school, Jounwl of Creative Behavior, 16(1). 50-57. Refernwes 363

DAVIS, H. B... &CONNELL. J.P. (1985). The effect of aptitude and achievement status on the self-system. Wird Child Quarterly. 29(3), 131-136. Dasis, J. 0929). A study of 183 outstanding communist leaders-item:4a Sociological Review. 24, 42-55. DAVIS, P., & HERSH, R. (1981). The mathematical experierwe. Boston: Houghton Mifflin. DE GROOT, A. D. (1978). Thought and choice in chess (2nd ed.). The Hague: Mouton. Dt LEON, J. (1983). Cognitive style differences and the underrepresentation of Mexican Americans in programs for the gifted. journal for the Eduration of the Wted, 6(3),167- 177. Driist.E, J. R. (1986). Death with honors:Sukt 4eamong gifted adolescents. Jonriuü of Counsel- ing and Develotnnent, 64(9), 558-560. DENBOW. T. (1984). Rhodes recipe. MW Today, 3(3). Durrscn, C. (1968). Environment and perception. In M. Deutsch, I. Kati, & A. Jensen (Eds.). Social class, nue, and psychological development (pp. 58-85). New York: Holt, Rinehart & Winston. DEWEY,J. (1902/1956). The Mild and the curriculum. Chicago: University of Chicago Press. DEYousw, A. (1987). The status of American rural education research; An integrated review and commentary. Review of Edurational Research, 57(2), 123-148. HEURLE, A., MELLINGER, H & HAGGARD, E. (1959). Personality, intellectual, and achieve . ment patterns in gifted children. Psychological Monographs, 13, 1-24. DLMITROVS/LT, L. & &am M. (1975). Early conservation as a predictor of arithmetic achieve. mem. Journal of Psychology, 91, 65-70. DOSZHANSKT, T.. & MONTAGU, A. (1975). Natural selection and the mental capacities of mankind. In A. Montagu (Ed.). Race and /Q (pp. 1(4-113). London: Oxford University Press. )OLAN, L. (1983). The prediction of reading achievement and self-esteem film an indexof home educational environment; A study of urban elementary students. Measurrim-nt and Evaluation in Guidance, 16(2) 86-94. DOMAN, G. j. (1964). How to teach your baby to read: The gentle revolution. New York: Random House. DOWDALL, C.. tic CLHANGEIO, N. (1982). Underachieving gifted students: Reviews and implications. Wird Child Qtuithily, 26(4), 179-1144. DREW% E. M. (1961). A critical evaluation of approaches to the identification of gifted students. In E. Arthur (Ed.), Measurement and research in today's school (pp. 109-12 1). Washington. IX:: American Council on Education. Dut.e.E. R. (1983). Interview with John "oh. Aegis: Newsletter of the WVIVA, 2(1),6-8. Charleston, WV: West Virginia Gh.ed Education Association. Available from Educa. tion Department. University of Charlestcni, Charleston, WV DUNCAN.A. (1969). Behavior rates of gifted and elementary school children (Monograph). Citwin. nati, OH: The National Association for Gifted Children. DUNCAN: 1. (1927). My hfe New York: 1sveright. DUNCUM. P. (1985). Bow 35 4 hildren born between 1724 and 1900 karned to clraw, Studies in Art Education. 26(2). 93-102. Deism, L., &MARKWARDT, F. (1970).Peabody individual cu hievenwat reg. Circle Pines, MN: American Guidance Service. Dupuis, M., & CARTWRIGHT, G. (1979, April). The writing of gifted children. Paper presentedat the American Educational Research Association Annual Conference. San Francisco, CA. DURDEN, W. (1980). Gifted progra.ns: '1 he johns Hopkins program for verbally gifted youth. Ruler (inflow, 43). 34-37. DURKIN, D. (1966). Children who read early. New York: Teachers College Dealt, W. K. (1981. January). Reading and the gifted student. Paper presented at the Annual Weting of the Southwest Regional Conference of the International Reading Associa- tion, San Antonio, TX. Eccus. J. (1985). Model of students' mathematics enrollment decisions. Educational Shobes in Mathematics. 16. 311-314. Foci:, 0., & FRIEDBERG, S. (1984). Factors affecting achievement in the first course in calculus. Journal of Experimental Education. 52(5), 136- 1411 EDMISTON, R. (1943). Do teachers show partiality? Peabody Journal of Education,20. 234-238. 372 364 References

EDMONI*, R. (1979). Effective schools for the urban poor.Educational Leadership. 37(1). 15-18. EDUCATIONAL TESVI1NG SERVICE. (1926-1987). College entranceexamination board sclwlastre aptitude test Princeton, NJ: Educational Testing Service. EDUCATIONAL TESTING SERVICE, COOPEI1ATIVE TESTDIVISION. (1956-1972), Sequential tests of educational progress. Princeton, NJ: Educational TestingService. EPSTEIN. C. B. (1979). The gifted and talented: Programsthat work. Arlington, VA: National School Public Relations Association. ERLICH, V. Z. (1979). The Astor program for giftedchildren prekindergarten through thr primary grades, five years later-Evaluation and problems of transitwn.(ERIC Document Reproduc- tion Service No. El) 171 054.) EHRLICH, V. Z. (1982). G#ted children: A guide ,for parentsand trachm. Englewood Cliffs. NJ: Prentice-Hall. EVSENCR. H. J. (1983). The roots of creativityCognitive ability or personably trait? Roeper Review, 5(4), 10-12. EYSENICK, H. J.. & EVSENCK, S. B. G. (1975). Maniud ofthe Evsenck personality cpwstwnruure. San Diego, CA: Edits. YEARN, 1,.. (1981. March-April). Writing: A basicatul developmental skill for gifted learners. GIC/T, 26-27, Fk.aitx. L. (1982). Underat hievement and rate ofacceleration. Gifted Child Quarterty, 26(3), 121-125. Eris, 0. (1910). Studien Uber die Genealogie und Psychokgieder Musuters. Weisbaden. Germany: Bergmann. FELDHUSEN, J. F., & Kt-a:ism:ma, H. J. (1962). Anxiety,intelligence, and achievement in children of low, average, and high intelligence. ChildDevelopment, 33, 403-409. FELDHUSEN, J.. & KOLLOF. M. (1979). An approach to careereducation for the gifted. Roeper Review, 2(2), 13-17. FELDMAN, D. (1979). The mysterious case of extremegiftedness. In A. H. Passow (Ed.), The gifted and talented; Their education and de velopmenl. TheSeventy-Eighth Yearbook of the National Society for the Study of Education, Part II (pp.335-351). Chicago: University of Chicago Press. FELDMAN, D. (1984). A follow-up of subjects scoringabove 180 IQ in Trrman's "'Genetic Studies of Genius." Exeeptional Children, 50(6), 518-523. FEVERSTEIN, R., MILLER, R., HOFFMAN, M. B., RAND, Y..MINTZKER, V.. & JENSEN. M. R. (1981). Cognitive modifiability in adolescence: Cognitive structuresand the effects of intervention. Journal of Special Education, 15(2). 269-287. FINE. B. (1967). Underachievers: How they can behelped. New York: 1)utton. FINK, M. B. (1962). Self-concept as it relates to academicunderachievement. Californm journal of Educational Research, 13, 57-62. FisnEa, K. (1987, November), Cognitive developmentin real children: Levels ana variations. Paper presented at Teaching Thinking and At-Risk Students, anInvitational Cross-Labora- tory Conference, Philadelphia, PA. FISHER, R. (1973). Muswal prodigies, masters at an early age.New York: Association Press. FLAVELE, J. H. (1977). Cognitive development. EnglewoodCliffs. NJ: Prentice-Hall. FussING, E. (1985). Career preparation. In R. Swassing(Ed.). Teaching gifted children and adolescents (pp. 340-375). Columbus, OH: Merrill. FONTEYN, M. (1979), The magk of dance. New York: Knopf. FORNESS, S. R., SINCLAIR, E., & GUTHRIE, D. (1983).Learning disabilities discrepancy formulas: Their use in actual practice. Learning DisabilityQuarter4, 6, 107-114. FOSTER, W. (1981). Leadership; A conceptual frameworktor recognizing and educating. Gifted Child Quarterly, 25(1). 17-25. FOSTER. W. (1985). Helping a child toward individualexcellence. In J. Feldhusen (Ed.), Toward excelience in gifted education (pp. 135-162). Denver, CO:Love. FosTER, W. (1986). The application of single subject researchmethods to the study of exceptional ability and extraordinary achievement. Gyted ChildQuarterly, 30(1), 33-37. FOWEALS, J. G. (1930). Adjusting the curriculum to theneeds of superior pupils. Nation'r Schools, 5, 82-84, 85. Fox, L. H. (I 976a). Facilitating educational developmenta mathematically precocious youth. ln J. C. Stanley, D. P. Keating, & L H. Fox (Eds.).Mathematical talent: Discovery, description, and development (pp. 47-69). Baltimore, MD:Johns Hopkins 1.Iniversity Press.

r Reerersces 365

Fox, L. H. (19764 Sex differences in mathematical precocity: Bridging the gap.In D. P. Keating (Ed.), Inteikdual talent: Research and development (pp. 183-214).Baltimore. MD: Johns Hopkins University PTCSS, Fox, L. H. (1976c). The values of gifted youth. In D. P. Keating (Ed.),Intellectual talent: Research and development (pp. 273-284). Baltimore. MD: Johns HopkinsUniversity Press. Fox, L. H. (1978). Scientists and mathema.kians of the future. In Advantage:Disadvan'agrd gifted (pp. 47-52). Ventura, CA: National/State Leadership Training Institute on the Gifted and the Talented. Office of the Ventura County Superintendent of Schools. Fox, L (1983). Gifted students with reading problems: An empiricalstudy. In L Fox, L. Brody, & D. Tobin (Eds.), Learning disable& gifted children (pp. 117-139).Baltimore, MD: University Park Press. Fox. L. It. lc COHN, S. J. (1980), Sex differences in the developmentof precocious mathematical talent. In L. H. Fox, L. Brody. & D. Tobin (Eds.). Women and themathematical mystique (pp. 94-112). Baltimore, MD: Johns Hopkins University Press. Fox. L. 11., & DENHAM, S. A. (1974). Values and career interests of mathematicallyand scientifically precocious youth. In J. C. Stanley. D. P. Keating, & L. H. Fox (Eds.), Mathematical taleoU: Disrovery, description, and development (pp. 140-175). Baltimore. MD: Johns Hopkins Univerity Press. Fox, 1.. H.. PASTERNAK. S. 11.. & PEISER, N. L. (1976). Career-related interests ofadolescent boys and girls. In D. P. Keating (Ed.), Intellectual talent: Research and development (pp. 242-261). Baltimore. MD: Johns Hopkins University Press. Fox. M. N. (1981). Creativity and intelligence. Childhood Education, 57(4). 227-232. FRAVENHE1M, J. (1978). Academic achievement characteristics of adult males who were diagnosed as dyslexic in childhood. Journal of Learning Divabdines. I I . 47b-483. FREEDMAN, 1.), G.. & FREEDMAN, N. C. (1969). Behavioral differences betweenChinese-Ameri- can and European-American newborns. Nature. 244.1227. FREEMAN. J. (1979). Gifted children. Baltimore. MD: University Park Press. FREEMAN. J. (1983). Emotional problems of the gifted child. Journal of Child P.sychologyand Psychiatry. 24. 481-485. FRENcn. J. L, & CARDON. B. W. (1968). Characteristics of high mental ability dropouts. The Vocational Guidance Journal. 16(3), 162-168, FRIENts. R. (1939). Influences of heredity and musical environment on the scores ofkinder- garten children on the Seashore Measures of Musk al Ability journalof Applied Psychol- og y, 27. 347-357. Funis ma THE ADVANCEMENT OP EMI:A/MN. (1957). Thry went to college early (Evaluation Report No. 2). New York: Fund lot the Advancement of Education. Gsonvs. W. II. (1980). Learning disabilities and brain functwn: A neurophysiological approach.New York: Springer-Verlag. Gs. N. (1978). The scientific basis of the art of teaching. New York: Teachers College Press. Gat:c. N.. & BERLINER, D. (1975). Educational psychology. Chkago: Rand McNally. GA1.LAGHER.1 (1958). Peer acceptance of highly gifted children in elementary school. The Elementary School Journal. 58, 465-470. (;ALLAHER, J. (1975). Teaching tht gifted child (2nd ed.). Boston: Allyn & Bacon. GALLAGHER, J. (1985). Teaching the gifted child (3rd ed.). Boston: Allyn & Bacon. GALLAGHER, J., Si CROWDER, T. (1957). The adjustment of gifted children in the classroom. Exceptional Children. 23. 306-319. GALLVCI, N. T. (1988). Emotional adjustment of gifted children. Gifted ChildQuarterly, 32(2), 273-276. Gaurosi, F. (1869/1962). Hereditary genius. London: Fontana. GALTON. F. (1883). Inquiries into human faculty and its development. London:Macmillan. GARDNER, H. (1973). The arts and human development. New York: Wiley. GARDNER, H. (1982). Art. mind, and brain: A cognitive approach to creativity. NewYork: Bask Books, GARDNER, H. (1983). Frames of mind: The theory of multiple trattligences. New York: BasicBooks. GARDNER, J. (1961). Excellence. New York: Harper & Row. GAVG, M. (1984). Reading acceleration and enrichment in the elementarygrades. Reading Teacher, 37(4), 372-376. GETZELS, J. W. (1979). From art student to fine artist: Potential, problemfinding, and perfor- 3 7 et 366 Referetwas

mance. In A. H. Passow(Ed.), The gifted arid tat/flied: Their educaiwn anddevelopment (pp, 372-387). The Seventy-eighth Yearbookof the National Society for the Study of Edu- cation. Part 1. Chicago: Universityof Chicago Press. Grrzus, J. W., & CSIRSZENTSIDIALYI, M. (1968). Thevalue orientation of art students as determinants of artistic specialization and creativeperformance. Studies in Art Educa- tion, 10, 5-16. GETLELS, J. W., & CSIXSZENTNIIHALYI, M. (1976).Thr creative vision; A longitudinal study of problem finding in art. New York: Wiley. GErzus, J. W., & JACKSON, P. W. (1962). Creativityand intelligence: Explorations with OW students. New York: Wiley. ConsELLI, E. E. (1963). Intelligence and managerial success,Psychological Reports, 12, 898. GINzstlitc, E. (1971). Career guidance: W ho needs it, whoprovides it, who can imprave it. New York: McGraw-HilL Gntoux, H. A. (1983). Theory and resistance ineducation: A pedagogy for the opposition. South Hadley, MA: Bergin Garvey. GLAss, G. (1983). Effectiveness of special education.Policy Studws Review, 2 (Spec.al 1), 65-78. GLIDEWELL, J., KANTOR. M., &sant, L., & STRINGER, L. (1966).Socialization and social struc. tore in the ck-ssroom. In L. Hoffman& M. Hoffman (Eds.), Review of child development research (pp. 221-256). New York: Russell Sage. GOERTZEL. V., & GOERTZEL, M., (1962), Cradles of eminence.Boston: Little, GoLo, M. (1965). Education of thr intellectually gifted.Columbus, OH: Merrill. GOOD, T., & BROPHY, J. (1987). Looking in clas.sroorris(4th ed.). New York: Harper & Row. GOODNOW, J. J. (1984). On being judged "intelligent."International Journal of Psychology, 19(4- 5), 391-406. GOODWIN, W. L., & Dinscou.. L A. (1980). Handbookfor measureinera and evaluation in early childhood education. San Francisco: Jossey-Bass. GoavoN, E. E. (1965). Music aptitude profile. Boston:Houghton-Mifflin. GORDON, E. E. (1979). Primary measures of musicaudiation. Chicago: GM. GORDON, E. E. (1980). The assessment of music aptitudesof very young children. Gifted Child Quarterly, 24(3), 107-110. CoutioN, W. J. (1961). Synergies: The development of creativecapacity. New York: Harper & Row. Gum, K. 0., & GOTZ, K. (1973), Introversion-extraversionand neuroticism in gifted and ungifted art students. Perceptual and Motor Skills, 36,675-678. Cknz, K. 0., & GITI7.. K. (1979). Personalitycharacteristics of professional artists. Perceptual and Motor Skins, 49, 327-334. GouLD, S. J. (1980). Women's brains. In S. J. Gould(Ed.), The panda's thumb (pp. 152-159). New York: Norton. GouLD, S. (1981). The mismeasure of man. New York:Norton. GOVI.DNER. A. W. (1979). The future of intellectuals and the riseof the new class. New York: Oxford University Press. GowAst, J. (1978). The facilitation of creativity through meditationalprocedures. Journal of Creative Behavior, 12(3), 156-160. GOWAN. J.. & DEstos, G. (1964). The educationand guidanee of thr ablest. Springfield, IL; Thomas. GRAY, R. A., SASKI, J., MCENTIRE, M. E., & LARSEN,S. C. (1980). IS proficiency in oral language a predictor of academic success? Ekmernary SchoolJournal, 80(5), 261-268. GREEN. W. W. (1985). Hearing disorders. In W. H.Berdine & A. E. Blackhurst (Eds.). An introdurtism to special education (2nd ed., pp. 183-234). Boston:Link, Brown. GREENFIE.LD, P. (1972). Oral or written language: The consequencesof cognitive development in Africa, the United States, and England. Languageand Speech, 15, 169-178. GRIFFIN, L., KALLEBERG, A., & ALEXANDER, K. (1981).Determinants of early labor market entry and attainment: A study of labormarket segmentation. Sociology of Education. 54(3). 206-221. GRIEnTH, D. R., & CLARK, P. M. (1981). MotiNation,intelligence, and creative behavior in elementary school children of low-creative ability. Journalof Experimental Education, 49(4), 229-234. Gaon, A. (1970). Genius in residence. Englewood Cliffs,NJ: Prentice-Hall. C.:ROTH. N. (1975). Mothers of gifted. Gifted Child Quarterly,19(5), 217-222.

3-1" r),_ References 367

GUILFORD, A. M., SCHEUERLE, J., & SHONBVRN, S. (1981). Aspects of language development in the gifted. Gtfted Child Quarterly, 23(4), 159-163. GunsoRD, J. P. (1950). Creativity. American Psychologist, 5, 444-454. GIMFORD, J. P. (1959). Three faces of intellect. American Psychologist, 14, 469-479. GUILFORD, J. P. (1967), The nature of human intelligence. New York: McGraw-Hill. GUILFORD, J. P.. & CMUSTENSEN, P. R. (1973). The one-way relation between creative potential and IQ. Journal of Creative Behavior, 7(4), 247-252. GUSKEY, T. (1984). Intpleinenting mastery learning. Belmont, CA: Wadsworth. HAGGARD, E. (1957). Socialization, personality, and academic achievement in gifted children. The School Review, 65, 388-414. HAIER. R. J.. & DENHAM, S. A. (1976). A summary profile of the nonintellectual correlates of mathematical precocity in boys and girls. In D. P. Keating (Ed.), Intellectual talent: Research and development (pp. 225-241). Baltimore, MD: Johns Hopkins University Press, HALLER, E. j.. & DAVIS, S. A. (1980). Dues socioeconomic status bias thy assignmentof elementary school students to reading groups? American EducationalResearth Journal. 17(4), 409-418. HARNESS, B.. EPSTEIN, R., & GORDON, H. W. (1984). Cognitive profiles of children referred to a clinic for reading disabilities. Journal of LearningDisabilities, 17, 346-351. HARRIS. M., MORGENBESSER, S., ROTHSCHILD, j., & WISHY, B. (Eds.). (1960).lntroduction to contemporary civilizatiim in the West (3rd ed., 2 vols.). New York:Columbia University Press. HASKELL, S. H.. BARRETT, E. K., & "[ASTOR, H. (1977). The education of motor and neurologwally handicapped children. New York: Wiley. HATHAWAY, S. R.. & MEEHL, P. E. (1951). An mica for the clinical we of tlw Minnesota multipluow personality inventory. Minneapolis. MN: University of Minnesota Press. HAvrx, B. B., & FREEHILL, M. F. (1972). The gifted: Case studies. Dubuque. IA: Brown. HAYES, C. D., & KAMERMAN, S. B. (Eds.). (1983). Chddren of working parents: Expenenresand oukomes. Washington, DC: National Academy Press. HEATH, S. B. (1982). Questioning at home and at school: A comparative study. In G. Spindler (Ed.). Doing the ethnography of sclwoling (pp. 96-101). New York: Holt. Rinehart & Winston. HEID. M. K. (1983). Characteristks and special needs of the gifted student in mathematics. Mathematics Teacher, 76(4), 221-226. Hrt.sots, R. (1971). Women mathematicians and the creative personality. Journal of Consulting and Clinical Psychology, 36(2), 210-220. HENSON, F. 0. (1976). Mainstreaming thr gifted. Austin, TX: Learning Concepts. HERSRERGER, J., & WHEATLEY, G. (1980). A proposed model for a gifted elementary school mathematics program. Cy-it'd Child Quarterly, 24(1), 37-40. HERRNSTEIN, R. (1973). IQ in the mnitocmry. Boston: Little. Brown. HERSEY. J. (1960). The child buyer. New York: Knopf. HETHERINGTON, E. M., CAMARA, K. A., & FEATHERMAN, D. I.. (1981). Cognitive prrjormance. selwol behavWr, and achWvement of children from one.parrnt households (Report prepared for Families as Educators Team). Washington, DC: National Institute of Education. (ERIC Document Reproduction Service No. ED 221 780) Ilwatarr, E. H. (1980). The relationship of logical reasoning ability, oral language comprehension, and home experiences to preschool chldren's print awareness. Journal of Reading Behavior, 12(4), 313-324. Hu_parrn, G. (1966). Introduction to the gifted. New York: McGraw-Hill. HOCEVAR, D. (1979). Ideational fluency as a confounding factor in the measurement of originality. Journal of Educational Psychology, 71(2), 191-196. HOCEVAR, D. (1980). Intelligence, divergent thinking, and creativity. Intelligence. 4(1), 25-40. HOESTADTER, R. (1963). Anti-intellectualiSm in American qe. New York: Knopf. HOLLAND J. L. (1973). Making vocational choices: A theoiy of careers. Englewood Cliffs, N. J.: Prentice-Hall. HOLLAND, J. L., & ASTIN. A. W. (1962). The prediction of academic, artistic, scientific, and social achievement of undergraduates of superior scholastic aptitude. Journal of Edura. timid Psychology, 53, 132-143. 1).," .) I r) 368 Raferrower

HOLLAND. J. & RICHARDS, J. M. (A65). Academic and nonacademic accomplishment: Correlated or uncorrelated? Journal of Educational Psychology. 56, 165-174. HOLLANDER, L. (1978). Extemporaneous speech presented at the National Forum on the Arts and the Gifted, Aspen, CO. June In E. Larsh (Ed.), Someone's Primity. Denver, CO: Colorado State Department of Education. (ERIC Document Reproduction Service No. El) 181 663) Howsiowoant, L. (1926). Gjlied children: Thrir nature and ntiltUre. New York: Macmillan. HOLLINGWORTH, L. (1942). Children above 180 1Q. New York; World Book. HOLUNGSVORTH, L., & Cosa, M. V. (1928). Children clustering at 165 IQ and children clustering at 146 IQ compared for three years in achievement, In G. M. Whipple (Ed.). Nature and nurture: Their influence upon achievement, Yearbook of the National Society for the Study of Education, Part II (pp. 3-33). Bloomington, IL: Public School Publishing. Hoan. J. M., LOEHLITC J. C., & WILLERMAN, L. (1979). Intellectual resemblance among adoptive and biological relatives: The Texas Adoption Project. Behavior Genetics, 1.133-148. Ihnowrit, V. (1965). Musk education: The background of re.search and opinion. Victoria. Australia: Australian Council for Educational Research. HORODEZKY, B., & LABERCANE, G. (1983). Educational and psychologicalmensurement. 43(2). 657- 662. A. (1986). Gifted education and the spectre of elitism. Journal of Education. 168(1), 117-125. HowLEs. A., & Howtiv, C, (1985). A personal retold: Is acceleration worth the effort? Rapper Revww. 8(1), 43-45. HOWLEY, A., HOWLEY, C., & PENDARVIS, E. (1986). Teaching gifted children: Principles and strategies. Boston: Little, Brown. llowt.Ey, C. (1982). Relationship between structural class background and intelhgence. Unpublished master's thesis, West Virginia College of Graduate Studies, Institute, WV. Howt.Ev. C. (1986). Intellectually gifte. students: Issues and policy implicatWns. Charleston. WV: Appalachia Educational Laboratory. HOWLEY, C. (1987a). Anti-intellectualism in programs for able students (Beware of gifts): An application. Social Epistemology, 1(2), 175- 181. HOWLEY, C. (1987b). It's controversial, but "accelerafion" could bring gifted kids up to full speed. American School Board Journal, 174(6), 32-33, 40. Hon-, K.. & HEBELER J (1974). Career education for gifted and talented students. Salt Lake City, UT: Olympus. Hun-, V. E. (1978). Creativity and transactional analysis. Journal of Creative Behavior, 12(3), 202-208. Home, M. (1987, April 20). Statistical picture of special education shows commitment, but problems. Education Daily, pp. 1. 3-4. HUNT, E. B. (1978). Mechanics of verbal ability. Psychological Review, 85, 109-130. HvaT, J. WV. (1961). lntelhgenee and experience. New York: Ronald. Hufmtv. A. (1962). Island. NEW York: Harper & Row. JACKENDOEF, R.. & LERDAHL, F. (1981). A grammatical parallel between music and language. In M. Clynes (Ed.). Music, mind, and brain: The neuropsychology of music (pp. 83-118). New York: Plenum Press. JACKSON, P. (1981). Secondary schooling for the privileged few: A report on a visit to a New England boarding school. Daedelu.s. 110(4), 117-130. .JACOB, E. (1981). Dancing. Reading, MA: Addison-Wesley. JACOBS, J. C. (1971). Effectiveness of teacher and parent identification tif gif ted children as a function of school level. Psychology in the Schools. 8, 140-142, Janos, P. J., FuNG, H. C.. & ROBINSON, N. M. (1985). Self-concept, self-esteem, and peer relations among gifted children who feel "different." Gifted Child Quarterly. 29(2). 78- 82. JANOS, P. J., Maaw000, K. A., & ROBINSON, N. M. (1985). Friendship patterns in highly intelligent children. Roeper Review, 8(1), 46-49. Jasrax, J.. & JASTAK, S. (1978). Wide range achievement test. Wilminstiv ), DE: Jastak Associates. JENCKS, C., SMITH, M., ACLAND, H. BANE, M. J., COHEN, D.. .;INTIS, H., HEYNS, B.. & MICHELSON, S. (1972). Inequality: A reasses.sment of the effect of lannly and schooling in America. New York: Harper & Row. 377 Referent:, a 369

JENCKS, C., BARTLETT, S., (ORCORAN, M., CROUSE, J., EAGLESFIELD,D., JACKSON, G., MCCLEL- LAND, K MUESFR, P., OLNECK, M.,SCHWARZ, J., WARD, S., & WILLIAMS, J. (1979). Who get ahead? The determinant of economic success inAmerica. New York: Bask Books. JENKINS, M. 1). (1943). Case studies of Negro children of Buret IQ160 and above. Journal of Negro Education, 12, 159-166. JENKINS. M. 1). (1950). Intellectually superior negro youth:Problems and needs. Journal of Negro Education, 22(2), 322-332. JENKINS-FRIEDMAN, R. (1982). Myth: Cosmetic use of mukipleselec:ion criteria. GOcted Child Quarterly, 26(1), 24-26. JENNINGS, H. H. (1943). Leadership and isolation, New York:Longmans. Green. JENSEN, A. (1969). How much can we boost IQ and scholasticachievement? Harvard Education- al Review. 39, 1-123. JENSEN, A. (1973). Educability and group differences. New York:Harper & Row. JENSEN, A. (1980). Bias in mental testing. New York: Free Press. JENSEN, J. (1973). Do gifted children speak an intellectualdialect? Exceptional Children, 39(4), 337-338. JOHNSON, D.. & MYKLERUST, H. (1967). Learning disabilities.New York: Grune & Stratton. JOHNSON, P. (1965). Art for the young child. In W. R. Hastie(Ed.), Art education (pp. 51-85). The Sixty-fourth Yearbook of the National Society for theStudy of Education, Part 11. (:hicago: University of Chicago. JoNEs, L. (1971). Black mune. New York: MorrowPress. KALINOWSKY. A. G. (1985). The development of Olympicswimmers. In B. S. Bloom (Ed.), Developing lair:: ;,z. young people (pp. 139-192). New York:Ballantine Books. KAMIN, L. (1981). Separatel identical twins. In H. J. Eysenck & L.Kamin (Eds.), The intelligence controversy (pp. 106-113). NewYork: Wiley. KAMPHAUS, R. W., & REYNGLDS, C. R. (1984). Developmentand structure of the Kaufman Assessment Battery for Children. Journal of Special Education,18(3), 213-228. Kallov, R. C., JOHNSON, B. 'W.. & KANDY, K. W. (1980). Locusof control and self-concept in achieving and underachieving bright elementary students. Psychology in'he Srhools, 17(1), 395-399. KARNES, F. A.. & CHAUVIN, J. (1986). The leadership skills:Fostering the forgotten dimension of giftedness. GICIT, 9(3), 22-23. KARNES, M. B. (1978). Nurturing academic talent in early childhood.(ERIC Document Reproduc- tion Service No. 161 528) KATCHADOURIAN, H., & Bola, J. (1985). Careerism and intellectualism amongcollege student. San Francisco: jossey-Bass. KATZ, M. (1968). The irony of early school reform. Cambridge. MA:Hat vard University Press. KATZ, M. (1971). Class, bureaucracy, and schools. New York; Praeger. KAUFFMAN. J. M. (1977). Characteristics of children's behavior disorders.Columbus, OH: Merrill. KAUFFMAN, J. M. (1981). Characteristics of children's behaviordisorders (2nd ed.). Columbus, OH: Merrill. KAunmAN, J. M. (1985). Characteristics of children's behaviordisorders (3rd ed.). Columbus, 011: Merrill. KAUFMAN, A., & KAUFMAN, N. (1983). Kaufman assessment batteryfor children: Sampler manual. Circle Pines. MN: American Guidance Service. Kavat.E., K. A., & FORNESS, S. R. (1985). Learning disaNlity and thehistory of science: Para- digm or paradox? Reme-lial and Special Education, 6(4), 12-24. KEATING, D. P. (1974). The study of mathematically precociousyLuth. In J. C. Stanley. D. P. Keating, & L. H. Fox (Eds.), Mathematical talent: Discovery,description, and development (pp. 23-16). Baltimore, MD: Johns Hopkins University Press. KEATING, D. P. (1975). Precocious cognitive development at thelevel of formal operations, Child Development, 43, 276-280. KEATING, I). P. (1976a). Creative potential of mathematicallyprecocious boys. In D. P. Keating (Ed.), Intellectual talent: Research and dmelopment (pp. 262-272).Baltimore, MD: johns Hopkins University Press. KEATING, D. P. (1976b). Discovering quantitative precocity. In I). P.Keating (Ed.), Intellectual talent: Researrh and development (pp. 23-31). Baltimore, MD: JohnsHopkins University Press. 37t!, 370 References

KEATING, D. P. (1976c). A Piagetian approach to intellectualprecocity. In I). P. Keating (Ed.), Intelkctual talent: Research and development (pp. 90-192).Baltimore, MD: Johns Hopkins University Press. KEATING, D. P. (1980a). Four faces of creativity:The continuing plight of the intellectual underserved. GPed Child Quarterly, 24(2). 56-61. KEATING, D. P. (1980b). Th;nking processes in adolescence.In J. Adelson (Ed.), Handbook of adolescent psychology (pp. 211-247). New York: Wiley. KEATING, D. P., & STAsu..EY, J. C. (1972). Extreme measuresfor the exceptionally gifted in math and science. Educational Researcher, 1, 3-7. KELLY, M. P. (1985). Unique educational acceleration:The dilemma of John Stuart Mill and contemporary gifted youths. aped ChildQuarterly, 20(1), 87-89. KELLY, T. K., BULLOCK, L. M., & DYKES, M. K.(1977). Behavioral disorders: Teachers' percep- dons. Exceptional Children, 41(5), 316-318. KEYHART, J., KEPHART, C., & SCHWARTZ. G. (1974).A journey into the world of the blind child. Exceptional Children, 40, 421-429. KERSH, M., & REISMAN, F. (1985). Mathematics forgifted students. In R. Swassing (Ed.), Teaching gifted children and adolescents (pp. 136-180).Columbus, OH: Merrill. KHA INA, J. (1978). The creatively gifted child; Suggestionsfor parents and teachers. New York: 'image Press. KIMBALI B. (1952). The sentence completiontecnnique in a study of scholastic underachieve- m u. Journal of ConsultingPsychology, 16, 353-358. adolescence. A mencan Journal of KIMBALI , (1953). Case studies in educational failure during Orthopiychiatry, 21, 406-415. KING, M. (1981). Rum/ delfuenr, proneness: ltsrelationship to giftedness, ernorvniartacd support. and environmental availability. Unpublishedmaster's thesis, University of Wisconsin, Platteville. KIRK, B. (1952). Test versus academic performancein malfunctioning students. Journal of Consulting Psychology, 16, 213-216. KIRK, S. A., & GALLAGHER. J. J. (1983), Educatingexceptional children (2nd ed.). Bo...in: Houghton Mifflin. KIRSTEIN. L. (1983). Classic ballet: Aria of the aerial.In R. Copeland & M. t..ohen (Eds.). What is dance? Readings in theory and mtirism (pp.238-243). Oxford: Oxford University Press. KrrAno, M., & KIRBY, D. (1986). Gptd education: Acomprehensive view. Boston: Little, Brown. KLAUSMEIER, H. J.. & LOUGHLIN, L. T. (1961).Behavior during problem solving among children of low. average, and high intelligence. Journalof Educational Psychology, 52, 148-152. KLEIN, R. D. (1982). An inquiry .nto the factorsrelated to creativity, Elementary School Journal, 82(3), 256-265. KLEINSASSER, A. M. (1986, October). Explorationsof an ambiguous culture: Conflicts facing gifted females in rural environments. Paper presented atthe annual sonfrrence of the National Rural and Small Sclu., 's Consortium, Bellingham,WA. (ERIC Document Reproduction Service No. ED 278 522). Kocn, K. (1970). MO:es, hrs, and dreams. New York:Vintage Books. KOCH, K. (1973). Rose, where did yini get that red? NewYork: Vintage Books. KOESTLER, A. (1964). The WI of creation. New York:Macmillan. KOHL, H. (1967). Thirty-six children. New York:Signet Books. KoHN, M. L. (1977). Class and ronforntity: A study invalues (2nd Ed.). Chicago: University of Chicago Press. KORTEN, D. C. (1972). Situational determinants ofleadership structure. In G. D. Paige (Ed.), leadmhip: Readings for an emerging field (pp. 146-164). NewYork: Free Press. KORZENIK, D. (1981). Is children's work art? Somehistorical views. An Education, 34(5), 20-'24. KRAUS, R. (1969). thstory of dance in an and education.Englewood Cliff, NI: Prentice-Hall. KRUTETSKIL V. (1976). The psyclwlogy of mathematicalabilities in schookhildren. I, Wirsrup & J. Kilpatrick, (Eds.). J. Teller (Trans.). Chicago: Universityof Chicago Press. (Original work published 1968) Kuaiszyst, T & BOIUCH, G. (1984). Educational Ic5ting and measurement;Classroom appliccaiim and practice. Glenview, IL: Scott, Foresman.

3 79 References 371

Roux. J., & Kum. C. (1984). Effects of accelerated instruction on students. RevWw of Edna:- &mai Research, 54(3), 409-425. KURTZMM, K. A. (1967). A study of school attitudes, peer acceptance. and personality of creative adolescents. Exceptional Children, 14, 157-162. Kovt.Estcy, W. P. (1973). Rural youth: Current status and prognosis. In D. Gottlieb (Ed.). Youth in contemporary sociey (pp. 321-345). Beverly Hills. CA: Sage. KWALWASSER, j. (1955). Expkring the musical mind. New York: Coleman-Ross, LABERGE, D., & Satatm.s, J. (1974). Toward a theory of automatic information processing in reading. Cognitive Psychology. 6, 293-323. Lanov, W. (1972). The logic of nonstandard English. In P. P. Giglioi (Ed.), Language and social context (pp. 179-216). New York: Penguin Books. LACHAPELLE, .1R. (1983). Creativity research: Its sociological and educational limitations. Studies in Art Education, 24(2), 131-139. LANE. V. (1960). The psychology of the actor. New York: John Day. LAW:WAY. L. (1983. March 28). Bringing up superbaby. Newsweek, pp. 62-68. LAWLER. L. B. (1964). The dance in ancient Greece. Middletown, Cl': Wesleyan University Prexs. Laveocx, F. (1979). Gtfted children. Glenview, IL: Scott,. Foresman. Um-ascii, E. (1969). Teaching and learning in city schools. New York: Basic Books. LEE, E. S. (1954 Negro intelligence and selective migration: A Philadelphia test of the Klineberg hypothesis. American Sociological Review, 16, 227-233. LEHMAN, E., & ERDWINS, C. (1981). The social and emotional adjustment of young. intellectually gifted children. (4ted Child Quarter/y. 25(3), 134-137. LEHMAN, N. (1986). The origins of the underclass. The Atlantic 237(6), 31-43, 47-55. LEITER, J., & BROWN, j. S. (1985). Determinants of elementary school grading. Socudagy of Education, 58, 166-180. LEONTIEF, W. (1982). The distribution of work and income. Scientific Amerman, 247 (3), 188-- 204. LESCOLD, A. M., & PERFETTI, C. A. (1978). Interactive processes in reading comprehension. Discourse Processes. 1 . 323-336. LESSER, C., DAVIS, F. B., & NAHLMOW, L. 9962/. The identification of gifted elementary school children with exceptional scientific talent. Educational and Psyrlwlogical Measure- ment, 22(2), 349-364. Lyssyr.. G. S., FIFER, G.. & Ctana. 1). H. (1965). Mental abilities of children from different maial-dass and cultural groups. Monographs of the Society for Research in Child Develop- ment, 30(4). Lywis, (:. L.. & KANES, L. G. (1980). Gifted 1EPs: Impact of expectations and perspectives. In 1). M. Jackson (Ed.). Readings in curriculum development f ur the giftrd (pp. 70-76). Guilford, CT: Special Learning Corporation. Lywis. R. B., k Dooniatt, D. 11. (1983). Teaching ler& students in the maimtream. Columbus, OH: Merrill. LEWONTIN, R., ROSE, S., & KAM1N, L (1984). Not in our genes. New York: Pantheon Books. LINES, P. (1986). Home instnoiwn: An overview. Charleston, WV: Appalachia Educational I..abo- ratory. LiPSET, S.. & DORA'S, R. (1972). The intellectual as critic and rebel: With special reference to the United States and the Soviet Union. Datdalus, 101, 137-198. lit eisica.ii, W., & SCHAEFER, T. (1984). Leadership through followership. In W. E. Rosenbach & K. L. Taylor (F.ds.), Contemporary 114sues in lradenhip (pp. 138-143) Boulder. Cfl: Westview Press. LORAN. W. (1967). The language Gf efrmentary sclwol children. Champaign, IL: National Council of Teachers of English. LOMBROSO, C. (1894 Thr man of genius. London: Walter Scott. LOPATE, P. (1978). Children, productivity, and creativity: The artist in the schools. Social Polio, 9(2). 42-47. LOWENFELD. V., & BRITTAIN, W. L. (1970). Creative and mental growth (5th ed.). New York: Macmillan. Loma, L. (1964). lndependence-conformity behavior as a function of intellect: Bright and dull children.ErreptionalChildren, 31(1), 5-13. :61) 372 ROirenees

LvoN, H. C. (1971). Learningto feel: Feeling to learn.Columbus OH: Merrill. MAccoav, E. E., & JAcRuN, C. N. (1974).The psychology of sex tittference. Stanford, CA: Stan- fordUniversity Press. MACINTOSH, D.K., & DUNN, L M. (1973), Children with majorspecific learning disabilities. In L. M.Dunn (Ed.). Exceptional children in the schools: Specialeducation intransitwn (2nd ed.). New York: Holt, Rinehart & Winston. MACKINNON, D. W. (1975).IPAR's contribution to thr conceptualiration and study creativity. In I. A. Taylor & J. W. Getzels (Eds.),Perspraives in creativity (pp.60-89). Chicago:

MACKINNON, D. W.(1978).In search ofhumaneffectiveness: Identifying and developing rreativity. Buffalo, NY: Creative Education Foundation. talent. In W. B. Barbe & MAcKINNON,D. W. (1962/1981). The nature and nurture of creative J. S. Renzulli (Eds.),P s y c h o l o g y and e d u c a t i o n t h e g i f t e d(3rd ed., pp. 11 1-127). New York: Irvington. (Original at dde published 1962) MACLEIsH, R. (1984). Gifted by nature, procligies arcstill mysteries to man. Smithsonian. 11(12), 70-79. MADDI,S. R. (1975). The strenuousness of' the creativelife. In 1, A. Taylor & J. W. Getzels (Eds.), Pmpectivesin creativity(pp. 173-190). Chicago: Mdine. MADDOX, C. D.. ScHEIBKR, 1.. M.. & BASS, J.E.. (1982). Self concept and social distance in gifted children. GqiedChildQuarterly, 26(2), 77-81. MAINts. D. (1985). Preliminary notes on a theory ofinformal barriers fOr women in mathematics.Educational Studies in Mathematics.16(3), 314- 317, MAKER, C. J. (1982). Teachinfe-learning models forgiftrd education Rockville,MD: Aspen Sys- tems. MAKER, C. J., RWDEN, M. R., TONELSON.S., &HOWELL, R. M. (1978). The self-perceptions of successful handicapped mu-mints. Albuquerque. NM: Universityof New Mexico, Depart- ment of Special Education. MANN. D.(1986). Thinking about the undoable: Dropout programs. In P.Penn(Ed.), Children at ?Ult.. An I trban Education Network conferenceproceedings(pp. 1- I l). Charleston. WV: Appalachia Educational Laboratory. MARcusr, H. (1978).The aesthetir dimensionBoston: Beacon Press. MARjoRismocs, K. (1972). ischievement orientation ofCanadian ethnk groups.Alberta Journal of Edurationd Research, 18(3), 162-173. A regression MARTORIBANKS. K.(197(i). Academk achievement, ;otelligence. and creativity: surfaceanalysis. Multivanate Behavioral &mart& 11(1), 105-118. MARK, M. (1978).Contemporary musu education.New York: Si hirmer Books. United States hy MARIANn,S. (1972).Eduration nf the gifted and talented: Report to the Cimgress of the the U.S. Commwmier of Education.Washington. DC: U.S. Government Printing Office_ MARLAND. S. (1974).Career edueation: A proposal for reform.New York: McGraw-Hill. of gifted and talented adoleswns: MARSHALL.B. (1985). Career derision-making patterns Implications for career education. journa/of Career Education.7(4), 305-310. Englewood MARTINSON.R. (1968).Curriculum enrichment for the gifted in the pnmars grades. Cliffs, NJ: Prentice-Hall. MAR rissoN, R. (1972). Background papers. In S. Marland'sEducationofthe gifted and talented (Vol. 2) Washington, DC: U.S. Department of Heahh. Education,and Welfare. MARZANO,R., & HUTCHINS, C. (1985). Thinking skills: A conceptualframework. A special issueof NoteworthyAurora,CO: MidcontinemRegional Laboratory. MAsLow. A. H. (1954). Motivation and prrsonahty. NeWYork: Harper & Row. MAl'HER, N., & UDALL A. J. (1985). The identificationof gifted underachievers using the Woodcock-Johnson Psycho-Educational Battery.Ruttier Review.8(1). 54- 56. MAY.F. (1982).Reading as unnmuniration. Columbus.OH: Merrill. &noir f our- McAsm. M., &CLARKE,A. (1976). Parent-offspring resemblance in intelligence. nid of Psychology, 67, 243-273. MCCALL, M. W., JR.(1984). Conjecturing about creative leaders. In W. E. Rosenbach& R. I.. Taylor (Eds.).Contemporary issues in leadership (pp. 271-280). Boulder,CO: Westview Press. McCuTi.AND, D..ATKINSON. CLARK, R., & LOWELL,E. (1953).The achievement owtrve.New York: Appf-'on-f:entury-Crolts. 3'1 Referroces 373

Mc.CoNstm, F. (1973). Children with hearing disabilities. In L. M. Dunn (Ed.). Exceptional children in the schools: Special education m transition (2nd ed., pp. 351-412). New York: Hob, Rinehart. & Winston. Mc Cut:KEN, K. A. (1960). Accelerating the reading speed of sixth grade children. Exceptemal Children, 27, 27-28. McIKAYli, 1).(1966). The art of theatre. In S. J. Cohen (Ed.). The modern dance: Seven siatements oj belief (pp. 53-61). Middktown, CT: Weskyan University Press. MCKINNEY, J. D., & FORMAN, S. G. (1977). Factor structure of the Wallach-Kogan tests of creativity and measures of intelligence and achievement. Psychology in the &hoots. 14(1), 41-44. MCNEMAR, Q. (1964). Lost: Our intelligence. Why% American Psychologist, 19.871-882. MEDNHA, S. A. (1962). The associative basis of the creative process. Psychological Review, 69, 220-232. MyoNicx, S. A., (1962). 'the associative basis of the creative ss. Psychological Review, 69, 220.-232. MEIER, N. C. (1939/1966). Factors in artistic aptitude: Final ...wary of a ten-year study of special ability. In E. W. Eisner & D. W. Eckel- (Eds.). Reading.s in art education (pp. 105. 116). Waltham. MA: Blaisdell. MERCER, J. (1981). The system of multicultural pluralistic assessment: SOMPA. In Balancing the scale for the disadvantaged gifted: Presentatwas from the Fourth Biennial Natiimul Confer- ewe on Doadvaniaged Gifted/Talented (pp. 29-57). Ventura, CA. Ventura Ceunty Supet - imendent of Schools. MEsst,, 1.., CRANo, W.. MESSE, S., & RICE. W. (1979). Evaluation of the predictive validity of tests of mental ability for classroom performance in elementary grades. Journal of Educational Psychology, 71(2). 233-241. MESTRE. j.. & GERM:E. W. (1986). .4 study of tar algebra acquisition of Hispanic and ring& ninth graders; Research findings relevant 10 teacher training and classroom practice. Unpublished manusclipt. University ot MaFsat linsens: Amherst. Mill-Ala+. R. J. (1978). Divergent thinking "threshold effect"-Ig, age, or skill? Journal of Expenmental Education, 47(1), 4 -H. MEYER. J. (1977). 'Fhe effec ts of education as an instnutimr American imam! of Sociology,. 87(1). 55-71. MICHAEL, W. B. (1983). Manifestions of creatislwhaviors by maturing pantie ipants in the Study of Mathematically Precocious Youth. In C. P. Benbow & J. C. Stanley (Eds.), adenur precocity: Aspects of ab development (pp. 38-50). Baltimore, MI): Johns Hopkins University Press. Miu.s. D. 1.. (1986). Why do more boys than girls receive special education? Contemporan Education, 57(2), 104--106. MILLER, C., & LILEREN, V. (1986). Mathematical uleas (5th ed.). Glenview, IL: Scott. Foresman. Murya. R. V. (1(456). Social status and socioempathic differences among mentally superiin mentally typical, and mentally retardedhildrew Exceptwnal Children, 22, 114-119. MILLER. S. D. (1979). Wolfgang Amadeus who? The negkct of music literature. Musw Educa- tors Journal, 65(8), 50-53. MILts. C. W. (1959) The pouter elite. New York- Oxford University Press. MILNE. A.. MYERS, D., ROSENTHAL A., & GINSBURC. A. (1986) Single parents, working mothers, arid the educational achievement of sa hocil children. Sociology of Educalum. 59 (July). 125-139. MITCHE1T, I V. (1959). Goal-setting behavior as a function of self-acceptance. over- and underachievement and related personality variables. Journal of Educational Psychology. 30(3), 93-104. MITCHELL, P. 13, (1984 An advocate's glade to budding support for gifted and talented rdwanun. vashington, DC: National Association of State Boards of Education. Mrrcon..t.. R. (1979). Less than to -di ran say. Boston: Little, Brown. M ircstui., S. (1976, April). Pan iital perceptruns of their experiernes with due process 17/stiff ml education: A preliminary report Paper presented 11I the annual meeting of the American Educational Research Association. San Francisco. CA. (ERIC Document Reproduction Service No. El) 130 482) Mpius. J. (1926). Genius as a b ological problem. Eugenics limew. 17, 242-257. 3 2 374 References

MONSAAS, J. A. (1985). Becoming a world-class tennis player.In B. S. Bloom (Ed.),Developing Went inyoung pop!, (pp. 211-269). NewYork: Ballantine Books. Moose.. C., NELSON-PIERCV, C., Mitt, M., & FRNE. D. (1984).Precocious conservation in c ()mem: The solution of quantity tasksby inmquamitative strategies.fiturnut ui mental Child Psychology, 38. 1-6. Morewomen in education but men earn more, report says.11987. August). TheHurdingtim Herald Dispatch.p. 1)5. MORGENSTERN, S. (Ed.). (1956). Crimpateroh maw: An Gag:4(v. 4 corripowt writingsprim Palrdnna in Ctipland. NewYork: Pantheon Books. MORSE, W. C., Ctrrwc R. 1... & FINK. A. jI. (1964).Publu school classes for thr emonimally handicapped: A researchunalssU. Washington, IX:: Council for ExceptionalChildren. Mos.-TELLER, F., & MIflNIIIAN. a (Eds.). (1972).On equality of educational opportunity.New Ycirk: Random IIOUNP. MUNSINGER, H. (1975). adopted child's IQ: A crithal review.Psychologual Bulletin, 82. 623-659. MURPHY. J. (198(1). Conflict. consensus, and tommtuluation:An inwrin erne report fur the Ann Arbor Symposium on the applications of psvc hology tothe teaching and training of musk. MumEducators Jourrud. 66(7).SI -532, (1958). 4 4owth pouesses in musice..oc a anon, In N. I lenrs (Ed.), MoirI tin( eptiin mita- education(pp. 140--162). The Fifty-Seventh Yearbook of' the National Society fin the Study ot Educaticm. Part 1. Chicago: Vniversity of ChicagoPress. MYERS, J. '1'. 09821. I ietnisp'sericitv research: An overviewwith some implications fin prob. tem soking.Jmirrud tii Crennur Behavior. /6(3). 197-211. Mviattrusr. & BostrEs. B. (1960). Psvchorwmokigical learning disorders in thildr en. Archri,. ofPrchatnes, 77,247-256. NAcur P. (1982).Rum( educatum: In search of a better was.Bouldet.Co;%Vests iew Press. NamoNAL COMMISSION ON EXCELLEMI IN EDII.M ION. (1983). .4nation at risk: The imperative for rihicaturnal reform.Washington, DC: S. Department of Education. N ReRAI. DEVEIJIPSH.N I INNA tint. ( 1986).Toward a definition of rural and small scholdt, Bellingham, WA: National Rural Devekpment Institute. NI- oars, ('.. M. (1985). Behavior disorders, In W. H.Berdine & A. E. BLit khurst. (Eds.). An introductum 1,,speeial education (2nd ed.. pp.427- 468). Bosttni: tattle. Brtiwn. NtAVE1.1.. A.. & SIMON. II. (1972). //n).viripinfrkm udenig.Englewood NJ: cm e-11,111, NORTON, H. (198(1). Intertelatimiships among music aptitude.IQ. and ndiron conservation. Jourrud of Research in Maui Education. 2814).21)7-217. NIT. R., 8e. NVE. V. (197I)).Motu in he elementary trhool(3rd. ed.). F.intlygood Cliffs. NI: Prentice- Hall Omws. 111985).ikerfnug hat k: How %Hoods struc tnre inequahly.New I Liven. GI : Yale 1' myet sny Press. OIHN, M. (1968).The fulfillment of promise: Torty-year follino up id the Terman gifted group Stan. ford, CA: Stanford Unisersity ';'tess. OIHN, A. (1895). Gni, tedes gioods &Minn Gem de Intro I ?arum, modernes Patis. ER. j.. & KM... P. (Eds.), (1978)Language in education: letting Ow tests km wde%. Newburg. OrsoN. M. B. (1984). What do vim mean In spaliarfRoeper Rrilfit%0, 240-2-14. Onit. K. (1955).Music for children.Translated and adapted by A. Walter & H. Hall. New Yolk: Associated Musk Publishers. Ortrox. S. T. (1937).Reading.wnhug,and speech problems in children.New Yolk; Norton. (Namm, A. F. 11953).Applied imagination.New Yin k. Osoosm. A. F. (1963).Applied imagination(3rd. ed.). New York: Scribner's. 0551:5K. & ROHAN. 13. J. (1931).Enriching du' curriculum for gifted children: A book of guidance for edui Wilma( administrators and classroom teachers.New York: Macmillan. Owls:, E. (1978). Dyslexia-Genetic aspects. In A. Benton & 1).Pearl (Eds.), Thstevin.4n appraisal of current knowledge (pp.266-284). New York: Oxford University Press. Oxford English Dictumary(cnmpact rd.. Vols. 1-2). (1971), New York: Oxford t Itiversits hess. PAFFEstrAiu,Erc. R. S., & AsNs. I). P. (19(36), Chronk disease infOrmer t ulkge students. AuterunnJournal of Publu- Health, 56,1026 -1936, Pm.t, E. B. (1976). A historical step beyond 'Ferman. In 1). P. Keating (El.). tun! talent: 3 3 References 375

Research and devdopment (pp. 295.-315). Baltimore, MI): johns flopkins Univer suy Press. PAivio. M. (1971). Imagery and verbal prttefiVi. New York: Holt. Rinehart k Wthston. PARN}S. S. (1975). CPSI-A program for Kilanced growth. Journal af CreativeBehavior, 9(3), 23-29. PARNES, S. (1981). CPSI: The general system. In W. Barbe & J. Renzulli (Eds.).Pswholugy and education af the OW (3rd ed., pp. 304-314). New York: Irvington. PARTRILICL, E. (1934). Leadership among adolescent boys. TeachersCollegemu'tbiawn. to Edu- intim (No. 608). New York: Teachers College Ptess. 1'.,ssow, A.. Got.unun;. M., & TANNENBAUM, A. (1967). Education of thedaudvantaged. New York: Holt, Rinehart & Winston. PAs'Ingt. N. (1949). nature.nurturr iontrinfersy, New York: Cidumbia University.. Kings Crown Press. Pt:AMMAN. (1983). A theoretical model tiff creativity. Education. /(I3(3). 294-.305. PEGNA It), C. W., & BiRcii. J. W. (1959). Locating gifted childrvi in junior highschool. Exceptional Cluldren. 25(7), 300-304. Pt.NnAavis. E. (1983). The written language of gifted children. Unpublished dot totaldisser ration. Universitv of Kentucky, Frankfort, KY. PkRKIss. D. N. (1981). Thr irand':s brYt work Cambridge. MA: Harvard Uni nsits Press. Pt.stRv. L. (1984). Intelleawil irk in America. New York: Franklin Watts. PE.1101.1). R. G. (1978). Identification of musically talented students. In E. (Ec1.). one's priority (pp. 58-.64). Denver, CO: Colorado State Depart nwin of Education.(FRU: Doc umem Reproduction Service No ED 181 663) Prtrirt R. S. (1980). The school-based interpnifessional team: Recurring problemsand some possible solutions. ournal of School Psychology. 18(4). 388-394. PHIPPS. P. M. (1982). The LI) learner is often a bosWhy? Anulemu Tlwrapy, /70). 425 430 Prim lAnAs. J. (1983) Community and tallith ehange ill rural Appa/whia. Chat lestim,WV. Ap- palachia Educational Laboratory. PINIM.Era, P. Mcistilltc, ShytmsoN, A. V. (19621. Student factors in foreign language kenning: A review ot the literature. Modern Language Journal, 41,. 160- I 70, Pixotio. R. (1982). ifted undvrachirvers. Roeper Review. 4(41, IS. 21. PLosystAx, P. (n.d.). What tau be done tin rural gifted and talented c hildren andsouth; In ldeas for urtrantrurapfirditafented (pp. 71-87). Ventura, CA: National/Siate Leadership Training Institute on the Gifted and the Talented and :he Office of the Vcrittita County Superintendent of School.;. PouLANTiAs. N. (1974). 011W% III (iintempoom uipitaloan. London: VeisAi. PiassYv, S. 11955). Con, ernaig the nature andni me of genius. Si lertitu Monthly. ,Y1, 123 129. Prosst.v. S. (1(#7). Fordling accelerates, Fen sears after, fit201,,''ofC0141114111g 11ssrho11rin. 14(11,73-80. PkitHARD. A.. & TAYrou. J. (1980). .4ue4erating /mining. The U-Vf4fiUggi1I0n .ficla.souroln Novato, (A: Academie Therapy PRINGLE ,SI. K. (197(1)..412fr misfits. London: Longman. him, M. (1979). Inner-city students' perceptions of the gifted filurard for Mr' Frirriofrori of Ihr Gifted, 2(2). 99-105. Rsen. 1, B.. Gor instin,, M. L., & Psssow, A. H. (191th). Bright underachievri sNew Volk. Teachers College Press. RAsiors. M. (1972). The channeling colons. In M. Cal any (Ed.). St hiwhing in a iorporate 0(.1eil (pp. 23-35). New Yin lc: Das id Mc Kers. READ, Ii. 11959). ,4 concise &shim of modern /limiting. London: Thames & Hodson. }LAD, If. (1960). The third realm of education. In Mr (rentwe arts in Amen( aneducation- lhc highs and Burton lectures, Cambridge, MA: Harvard University Press. Rini, I). K.. Mass°. W. P.. & 11Asnalir.. 0. (1983). Test of early reading ahihn. Los Angeles: Western PSVL hological Services. Rriumt.. C. K. (1979). Genius creativrty and eminence. Gifted Child Quarterly. 27(4).837- 854. RENZU11.1.1. I. N. (1977).11ir enrichment triad 'nixie!: A guide for developing defensible programs for thc gifted and talented. Wethersfield. CF: Creative I,earning Press. Rirsit'vtr, J. S. (1978). What makes giftedness.! Phi Delta Is'aNuin. 60(3). 180 --184. 261. 376 References

RENZULLI, & HARTMAN. R. K. (1971). Scale lot rating the behavioralcharacteristics of superior students. Excepwrial Children, 38, 243-248. RENZUW. J. S., SMITH. L. IL, WHITE. A. 3., CALIAHAN. C. M..& HARTmAN, R. K. (1977). Scales for rat* the behavioral characteristics of superior students.Mansfield Center, CT: C:rearive Learning Press. RESNICK, L. & Foal), W. (1981). The psychology of mathematics for Distructwn.Hillsdale, NI: Lawreire Erlbaum Associates. RESTAK, R. M. (1979). The brain; The last frontwr. New York: WarnerBooks. REYNOLDS. C. R. (1985). Critical measurement issues in learningdisabilities. journal of Special Education. 4(4), 451-476. REVIEOLDS, C. IL, & RICHMOND, B. 0. (1985). The revised children'smanOt anxiety vale. 1.os Angeles: Western Psychological Services. Ripon.. F. S. (197)). Related aspects of the social and ffonomic problems,cultural tradition, and educational system of rund Appalachia: An analysis based on the concept ofscale. Unpublished doctoral disseration, Ohio State University. Columbus, OH. Roast. S.. & DAVIS, G. 0980). Five years of international researchwith GIFU: An instrument for the identification of creativity.fournd of Creative Behavior, 14(1), 20-24 THE ROPON SocIETE. (1969). The les.sons of victory. New York:Dial Press. Rm. R. (1970). Student social class and teacher expectations: Theself-fulfilling prophcxy in ghetto education. liartord Educational ReVitlio, 40, 411-451. Rirvo, E. R. (1977). Biochemical studies of children with the ...,ndromesof autism. childhood schizophrenia, an 3 related devdopmental disabilities:review. Journal of Chdd Psycho/ op and Psychiatry, 18. 373-379. ROBINSON, N., & JANos, P. (1986). Psychological adjustment in acollege-level program of marked academic acceleration. Journal of Youth and Adolescence. 15(1), 51-60. Rot., A. (1943/1975). Painters and vainting, In I. A. Taylor & J. W. Getzels(F.ds.), Persp awes in creativity (pp. 137-172). Chicago: Rm. A. (1952). A psychologist examines 64 eminent scientists. SnentituAincti,an. 18";(5). 21, 25. RoE, A. (1953). The making of a scientist. New York: Dodd, Mead. RoE, A. (1955). The psycludogy of occupations. New York: Wiley. RousEu. W. C. (1978. August). Social deiielolmient in nattiertually advanced children Paper presented at the Annual Convention of the American PsychokigicalAssociation, Toromo. Canada. RoEDELL, W. C.. JAcKsoN, N. L. & ROBINSON. H. B. (1980). Gifted Noungchildren. New Vol k: Teachers College Press. RoErIER. A. (1966). Finding the due to children's thought pox esses.Young Children. 21. 335- 34ti. ROITER, A. (1978), Some thoughts about Piaget and the younggifted child. Gifted Cluld Quarterly, 22. 252-257. ROGERs. C. (1959), Toward a theory of creativity. In H.Anderson (Ed,), ernitivey and its cultivation (pp. 69-82). New York: Harper & Row. Ross.. I.. H.. & Its. Hs.T. (1984). A meta-analysis of long-term ceativity training programs. Journal of Creative Behavior. 18(1), 11- 22. RosE, R. (1979). A program model for altering children's consciciusness.(;ifted Child Quarterly. 2(1), 109-117. ROSE. S. J. (1986). The Amernan profile poster: Who wowwhat, who makes how much, who works where, and who lives with whom. New York: Pantheon Books. ROSENBAUM, J. (1975). The stratification of the socialization process. AnurrranSorwlivral Review, 40 48-54. RoSERMALD, P. J. (1984). Breaking away '80's style. Dancemagazine, 58(4)70-74, ROTH, R. M.. & Pon. P. (1967). Direction of aggression andthe underachievement syndrome, Journal of Counseling Psychology, 14(3). 277-281. Rua-awes, P., & MAEHR, R. (1973). iNgmalion black and white, Journal ofPersonahry ond Social Psyvholop. 2,5(2), 210-218 Restaraofx. R. (1984). The joh market for college graduates, 1960-1990.1ournal of Higher Education, 55(4), 433-454.

3 5 References 377

York: Vhitage Books. Ryar. W. (1976). Blaming the Maim. New 97-121). RYSER. C. P. (1964). The student dancer. InR. N. Wilson (Ed.). The aru in society (pp. Englewood Cliffs. NJ: Prentice-Hall. instructing the gifted in art. Art Education. 27(3),16- 19. SALOME, P. A. (1974). Identifying and (3rd ed.). SALVIA, J.. & VSSELDVia. J. E. (1987).Amessment in special and remedial education &num: Htiughton Mifflin. giftedness. Gifted Child Quarterly. 2812). 95-96. SAUER. R. T. (1984). Early reading and New SASIUDA, R. J. (1975). Pywhologu al testing ofAmerican mmonties: ISSUfi and echiverleriCh. York: Dodd, Mead. SAmEtt. J. M. (1982). Autumn, ofchildren% intelligence and special abilities (2nd ed.). Boston: Allyn & Bacon. SAUNDERS. P. J. (1982). Sc. reviling and identifyingthe talented in art. Roeper Reline, 4(3). 7- 10. SAWYER, R. N. (1988). In defense of acadernkAgor.journalfor. the Education cif the Gifted, 11(2). 5-19, SCARR, S. (1981). Rare, soda/ c/asy, andmdrodual differences in /Q. Hillsdale. NI: Lawrence Erlbaum Associates. white SCARR, S., & WEINBERG, P. (1976). IQ testperformance of black children adopted by families. American Prwhologut 31, 726.-739. ScHLOSSER. 1... & ALGOEE1NE. B. (1980). Sex.behavior. and teaeher expectancies. Journal of Experimental Education. 48, 231-236. Schm.wnmxi, E. & REYNotsts, C. R. (19145).Dimensions ol anxiety among high IQ children. G4ted Child Quarterly. 29(3), 125-130. SCHONBER(.. H. C. (1970). At Ryes of the gtrat composer.t.New York: Nmuni. Journal SCHUBERT. D. (1973). Intelligence as necessarybus not sufficient for creativity. Genetic Psychology, 122(1), 45-47. ranking low in Sciscw. R. E. (1958). Comparisonof Negro pupils ranking high with those educational achievement. Journal of EducationalSociology, 31,265-270. :mini, B. (1986). The declineof literacy and liberal learning. Journal Education, 168(1). 105-116. arr, E., & BREAM. B.(1978).Sot ial interactions of early-reading andnon-reading kinder. garten students with highintellectual ahililty. Catalog of Selected Documents inPsychology. 8, 95. ScRIPTIAti, E. W. (1891). Arithmetical prodigies.American Journal of Psychology. 4, 1-59. .Si..surir.. M. V. (1975). 'Ferman and the gifted.Los Altos. CA: Kaufmann. SEARS, P.. & %man:. A. (1977). Careerand life satisfactions among Terman's gifted women. In J. Stanley. %4'. George, & C. Solano(Eds.). The gifted and the rreatne: .4 fifty-year perspective (pp. 28-65). Baltimore. MD:Johns Hopkins University Press. 1938) St ASHORE, C. (1967). Psyrhology of mon-.New York: Dover. (Original work published K. R. (1984). Perspectives on adokscentgiftedness and delinquency. Journal for tlw Education of the Gifted. 8(1). 59-72. in the St:on.., S., BessE.T., & MANSE11.1.1),R. (1980). The relationship of sdentific creativity biological sdences to predoctoral accomplishmentsand expel Unites. American Edura- tumal Research Journal. 17(4). 49 I -501. SELLS. L. W. (1976, February). The mathematicsfilter and the edocation of women and mmonties. Paper presented at the Annual Meeting ofthe American Association (or the Ad- vancement of Science. Boston, MA. (ERICDoc ument Reproduction Service No. E) 121 663) Silo. G. M. (1983). The nature and idemification Iflearning disabilities and their relationship to the gifted child. In L. H. Fox, L.Brody , & D. Tobin (Eds.), Learning duabled/grffrd children, Baltimore, MD: University Park Press. SESSIoNS, R. (1965). The musical experience of composer,performer, and /wrier. New York: Athe- neum. (Original workpublished 195(1) SIVHONS, R. (1979). Roger Sinuous on music: Collected fA.VIV3.Princeton, N J: Prim don Untvei sit% Press. Srutots, P. (1969). The feminized male:Clastinoms. white rallari, and the decline of mardntesi. New York: Random }/Ott%V. 378 Rdfenrysen

SHADE, B..1. (1978). Social-psychologkal tharacteria.cs of achieving black t Itihken. Negro Educational Review, 29(2), 80-86. SHADE, B. J. (1981). Personal traits of educationally successf ul black children. Negro Education- al Review, 32(2), 6-11. SHANTZ. C. (1975). The devekipment of social cognition. In E. M. Hetherington (Ed.). Review of child development nsrarch (Vol. 5). Chicago: University of Chicago Press. SHAW, M. C., & GRUM!, J. (1958). Hostihty and able high school athievers. jnurnal of Cournd- ing Psychology. 5(4). 263-266. SHAW. M. MCCULIN. J. 1. (1960). The onset of academic underachievement in bright children. Journal of Educational Psychology. 53(3), 103-108. SHER, J. (Ed.). (1997). Education in rural American: A leassessment of the conventional wisdom. Boulder. CO: Westview Press. SHERWOOD, J.. & NATAUPSKY, M. 09610. Predicting the conclusions of Negro-white intelligence research from biographical characteristics of the investigator, journal of Person. ality and Social Psyr-hology, 8(1). Part I. 53-58. SHrITEIL. D. (1985). Prenatal music experiences. Music Educators journal, 71(7). 26-27, Smut's, J. (1978). MI. twins: Their use and alnise. In W. Nmwe (Ed.). Twin research fisu !whip. and methodology. New York: Liss. Sinus. M.. WE:ATMS-S. D., HowAtto, L. & CivEtss. R. (1985. January 17). A portrait of America. Newsweek. pp. 20-33. SHILS, E. (1972). The intellectuals and the powers and other nsays. Chicago: I:nivel:say of Chic ago Press. SHNEIDMAN, E. (1..81). Suicide among the gifted. Suicide and 1.1r-Thoutening Behavior. 11(4), 254-282. SMITE, K. (1964)..4ninvedigatum of hereditary and environmental factors in musual abibty. published dot total dissertation. University id London. Smurfit, R. (1968). The psychology of nnoical ability. London: Methuen. SuNtit. L. A. (1971). A proposed view on tlw etiology of the neurological learning disability syndrome. journal of Learning Mallows. 4. 123-132. Soma, K.. & DEYouNti. A. (1986). The ideology of rural/Appalat hian education, 1895-1935: The Appalachian education problem as part of the Appalat Wan life probkm. Ethan- firma! Theory, 76(1), SILVERMAN, K.. & LANIER. V. (1965). Art fin the adolescent. 11) W. R. Ilastie (Ed.). Art education (pp. 115-152). The Sixty-fourth Yearbtxrk of the National Society tor the Study of Education. Chicago: University of Chicago Press. SIMON, A., & WARn, 1.. 0. (1973). The performance of high and low ability groups on two measures of creativity. journal of Experimental Eduratwn. 42(1), 70-73. SIMONTON, I). K. (1984). Genius. creativity. and leadership. Cambridge. MA: Harvard University Press. SIMLA, E.. & HAYDEN, S. (1965). Exploring eitktic imagery among the retarded. Perreptwil and Motor Shills. 21. 275-286. SISK, D. A.. & BiLio.v. K. ;1979). Every child in a gifted piograin? Irotruilor. 88(9). 941. 92. SUER. 1'. (1984). tiorare's compromise. Koslow Houghton Mifflin. SKODAK, M., & SKEELS. 11. (1P49). A final follow-up study of one hundred adopted children journal of Genet- Psythology, 75. 85-125. SLAVIN. R. (1986). Learning together. American Edurator: The Professwrud journal id the American Federation of Teachers, 10(2). 6- 11. Smart. C. R. (1983). Learning dualnlities: The interartwn of learner. task, and setting. Boston: Little, Brown. SMITH, S. B. (1983). The great menial calculators. New York: Columbia University Press. SMOLAK, 1. (1982). Cognitive precursors of receptive vs. expressive language. journal of (Judd Language, 9, 13-22. SNUGG, D. (1938). The relation between the intelligence of mothers and of their children living in foster homes. journal of Gemdir Psychology. 52, 401-406, SOLANO, C. H. (1977). Teat her and pupil stereotypes of gifted boys and girls. Talents and Gifts. 19, 4-8. SORREL.. W. (1967). The dance through the ages. New York: thosset & Dunlap. Referowes 379

SOSNIAK, L. (1985). Learning to be a cmwert pianist. .1 B. S. Blocm (Ed.), Developingtalent in Young .Peapk (pp. 19-67). NewYork: Ballantine. SPEARMAN, C. (1927). The abilities of man. New York: Macmillan. SPRINGER, S., & DetrINCH, G. (1981). 14? brain, right bruin. San Francisco: Freeman. STALKER, M. Z. 0081). Identification of the gifted in art. Studie3 en Art Education, 22(2) 49-56. STANISLAVSKI, C. (1948). My lift in art. New York: Meridian Books. STANLEY, J. (1976a. Septcr .r), Brilliant youlk Improving Thp quality and speed of their echo alum. Paper presented to the .t..ighty-fourth Annual Conference of the American Psychologi- cal Association. Washington, DC. (ERIC Document Repmduction Service No.ED 136 536) STANLEY, J. C. (1976b). Use of tests to discover talent. In D. P. Keating (Ed.). Intellectual talent: Research and development (pp. 3-22). Baltimore, MD: Johns Hopkins University Press. STANLEY, J. (1977). The predictive value of the SAT for brilliant seventh- andeighth-graders. The College Board Revww, 106, (Winter, 1977-1978), 31-37. STANLEY, J. (1981). Rationale of the study of mathematically precocious youth(SMPY) during its first five years of promoting educational acceleration. In W. Barbe & J.Renzulli (Eds.), Psychology and education of the gifted (3rd ed., pp. 248-283). New York:Irvington. STANLEY, J. C. (1985). A baker's dozen of years applying all four aspects ofthe Study of Mathematically Precocious Youth (SMPY). Rneper Review, 7(3), 170-173. STANLEY, J. C. (1986. April). Simw rhararteristies of SMPY's "700-800 an SAT-M before age17" group. Paper presented at the Sixty-Seventh AnnualConference cif. the American Edu- cational Research Association, San Francisco, CA. (ERIC Document Reproduction Ser- vice No. 277 206) SIANI.EY, J., & IIENBOW, C. (1983). SMPY's first decade: Ten years of posingproblems and solving them. The Journal of Special Education, 17(1), 11-25. STANLEY, J., KEATING, D., & Fox, L. (Eds.). (1974). Mathematical talent: Duroven, description. and development Balti4:icne, MD: johns Hopkins University Press. ANOvIcH, K. E.. CUNNiN MANI, A. E., & FE:Est:AN, D. J. (1984).Intelligence. wgilitive skills, and early reading progress. Rending Researeh Quarterly, /9(3), 278-303. STEDMAN, I.. (1924). Eduration of gifted children. Yonkers-on-Hudson, NY: WorldBook, STERN, U. U.. & CummINs. J. (1981). Language teaching/learning research:A Canadian perspective on status and directions. In J. K. Phillips (Ed.). Artwn for the 81is: Apaitk. pnle,imiat, and public program for foreign bmguage education (pp. 195-248).Skokie, IL: National Textbook. SI ERN. W. (1914).The psyrhologind methods of wing intelligence (G. M. Whipple, 'trans.). Bal- timore. MD: WwwWk & York. STERNBERG. R. (1977).Intelligence, information processing, and analogical reasoning: The r iimpourn- hal analysts of human abilities. Hinsdale. NJ: F.rlbaum. STERNBERG, R. (1981). Acomponential theory of intellectual giftedness. Gifted Chrld Quarterly. 25, 86- 93. STERNBERG, K. (1982a).Lies we live by: Misapplication of tests in identifying the gifted. Gi fled Child Quarterly. 26, 63-67. STERNBERG, R. (1982b, April). Who's inteHigenr, How the layperson definesinlelligerue. Psychology Tricky, 16. 30-33. STERNBERG.R. (1985). Beyond IQ: A truirrha theory of human mtelligerwe. Cambridge: Cambridge University Press. STERNBERG, K. & GARDNER. M. K.(1983).Unitk s in inductive reastming.journalq Expenmen tal Psychology: General, 112, 80-116. STEvENS,E.. & Woons. G. IL (Eds.). _justice, ideology, and educatnin: An introduction tothe social foundations (4 educatiim, New York: Random House, STocKARD, J., LANG, D.. & WOon, J. W. (1985), Academic merit, status variables,and students' grades. Journal of Reward; and Development in Eduration, 18(2), 12-20. STociotto, J., & MAYBERRY, M. (1986). The relationship between school environment andstuck:a achievement: A rePIFIV of the literature. Eugene, OR: Center for Educational Policy and Management. S loutau., R. (1974). Handbook ofleadership: A otrzev of theory nnd research. New York: Free Press. 3 8 3110 References

&row ism S. S., & LESSER, G. (1967). Learning patterns in the disadvantaged. Harvard Educational Revirw. 17.546-593. SeRActiev, J. (Ed.). (1961). Sigmund Freud's-Civilization and its discontents." New York; Norton. STRANG, B. (1954). Reading development of gifted children. Elementary English, 31, 35-40. STRANG, R. (1956). Gifted adolescents' views of growing up. journal of Exceptional Children. 23, 10-15. STRANG, R. (1963). Psychology of gifted children and youth. In W. Cruickshank (Ed.). Psychol- ogy of exceptional children and yotah (2nd ed.. pp.484-525). EnglewoodCliffs. NJ: Pren- tke-Hall. SexAms, A. A., & LermNEN, L E. (1947). Psychopathology and education cl the brain-onared New York; Grum: & Stratton. STROMEYER, C. (1970). Eidetikers. Psycludogy Today. 4(6) 76-80. Svc:Envoi, R. (1975). A model for the analysis of inquiry. In W. Barbe & J. Renzulli (Eds.). Psychology and eduration of the gifted (2nd ed., pp. 336-345). New York: Irvington. SULLUAN, E. (1967). Plage and the school curriculum: A critical appraisal Bulletin No. 2 of The Ontario Institute for Studies in Education. Toronto: Ontario Institute for Studies in Education. SUTARIA, S. 1). (1985). Specific learning disabilities: Nature and nerds. Springfield, IL; Thomas. SwARD, K. (1933a). Jewish musicality in Amerka. Journal of Applifd Psythology, 17(6). 675-712. SWARD, K. (1933b). Temperament and di; ection of achievement.fouinal of Social Psychology, 4, 406-429. SwAssiNG. R. H. (Ed.). (1985). Teaching gifted elnldren and adolescents. Columbus, OH: Merrill. Svrtmes. D. (1972). Gyted and talented children: Practical programnung for teachers and principals. Arlington, VA: The Council for Exceptional Children. TARA. H. (1965). Learning by discovery: Psychological and educational rationale. In J. J. Gallagher (Ed.), Tearhing gifted students: A hook of readings (pp. 177-186). Boston: Allyn & Bacon. TANNLNRAvm. A. (1962). Academn- attitudes toward academic brilliance. New York: Bureau ot Publications. Teachers College, Culumbia University. TANNENBAUM. A. (1981). Pre-sputnik to post-Watergate concern about the gifted. In W. Barbi. & J. Renzulli (Eds,). Psycholop and Mutation of the gifted (3rd. ed.. pp. 20-37). New York: Irvington. 'TANNENBAUM. A. (1983). Gifted children: Psychologual and edutatumal perspectives. New York: Macmillan, TAo, B. (1986). Parental involvement in gifted education. Educational Studies in Mathematics, 17, 313-321. 'TAYLOR. U. W.. & ELLISON, R. L. (1975). Moving toward working models in creativity: Utah reativity experiences and insights. In 1. A. Taylor & J..W. Getzels (Eds.). Perspectives in creativity (pp. 191-223). Chicago: Aldine. 'TAYLOR. J. (190/). TO do and not to see: The teacher of art. In E. W. Eisner & D. W, hker (Eds.). Readings in art education (pp. 238-245). Waltham, MA: Blaisdell. 'TAYLOR, M. L. (1975). Idea people. Chicago: Nelson-Hall. TEREERTILLER, C. (1986). Family characteristics of gifted versus non-gifted students. Graduate Monographs, 4( 1), 37-40. TERMAN. L. (1925). Mental and physital traits of a thousand gifted children: Gown- studies of genoo (Vol. I). Stanford, CA: Stanford University Press. 'FERMAN. L. M. (1973). Concept mastery test. New York: Psychological Corporation. TERMAN. 1. M.. & Muotru, M. A. (1973). Stanford-Binet intriligence male: Manual for thud revision (Form L-M). Boston: Houghton-Mifflin. TrnmAN. I.. M.. & OBEN. M. H. (1947). The gifted Adz/ grows up: Genetic- studiesgenius (Vol 4). Stanford. CA: StanfUrd University PITA& TERMAN, L. M. & ODEN, M. (1959). The gifted group at mui-lijr: Genetic studies of genius (Vol V). Stanford. CA: Stanford University Press. THORNMKE. R. (1972). Review of Torrance Tests of Creative Thinking, 1 n 0. K. Buros (Ed.), The seventh mental measurement yearbook (pp. 838-839). Highland Park, NJ: Gryphon. THORNDIRE. R.. & /1AC,EN. E. (1959). Ten thousand carters. New York: Wiley. THRAsnut, F. (1927). The gang. Chicago: University of Chicago Press, Inunow. 1.. ( 1987). A surge in inequality. Scientific American, 256(5), 30-37. References 381

THURSTONE. L. L. (1938). Primary menialabilaws (Psychonietrk Monographs No. 1). Chicago: University of Chicago Press. lied TiowEEE, R. (1980). A psycho-educationalprofile of 1,593 gifted high school students. Gs Child Quarterly, 24(2), 63-6F. TIMAR., T.,& Kinp, D. (1987). Educational reformand institutional cinupeteme. Harvard Educahonal Review, 37(3).308-1330. TISDALL., W. J.. BROWN, D. L., BYNUM, C.D.. & ROBINSON, S.(3972). Junior--Gifted and disadvantaged. In B. B. Hauck & M. F.Freehill (Eds.). The gifted-Case stuthes (pp. 86- 104). Dubuque. IA: Brown. Journal TONEMAH, S. (1987). Assessing AmericanIndian gifted and talented students' abilities. far the Education cf the G41ed.10(3), 181-194. P. (1962). Guiding cirstivetalent Englewood Cliffs. NJ: Prentice-Hall. ToRRANcE, E. potential. Minneapolis. MN: Universityof TORRANCE, E. P.(1963). Education and the creative Minnesota Press. behativr. Englewood Cliffs. NJ: Prentice-Hall. TORRANCE. E. P. (1965). Rewarding creative Princeton, NJ: Per- ToRitArgri, E. P. (1966). Torrance ,tat$of nrative thinking: Research edition. sonnel Press. thinking. Lexington. TORRANCE, E. P. (1974).Norm-technical manual: Torrance tesLY of creative MA: Ginn. Unique needs of the creativechild and adult. In A. li. Passow (Ed.), TORRANCE. E. P. (1979). Yearbook of The gifted and tafrnted: Their educationand development. The Seventy-eighth the Natitmal Society for the Studyof Education (pp, 352-371).Chicago: University of Chicago Press. longitudinal study. Creative 'TORRANEE. E. P. (1980).Growing up creatively gifted: A 22-year Child and Adult Quarterly. 5(3),14,1-158. 170. in identification of the gifted andtalented. TORRANCE,E. P. (1984). The role of creativity Gyled Child Quarterly, 28(4). 153-156. Creative leartUng and tenchmg. NewYork: Dodd, Mead. TORRANCE., E. P..& MvEns,R. E. (1971). Mathematics achievement in U.S. schools:Prelim- 'TRAVERS,K & MCKNIGHT. C. (1985). inary findings from the setondlEA mathematics study. Phi DeliaKappan, 66(6). 407-413. the gifted? Reading Tearher, 31, 742- TREZISE, FL (1978).What about a reading program for 746. Cambridge. MA: TvAcx. D. (1974), The one best .system: Ahisto7 of American urban edur-gion. Harvard University Press. mitruaion. The l'yEER, F. (1962). Intraindividualvariability. In N. B. Henry. individualizing Sixty-first Yearbook of the National Society for theStudy of Education (pp. 164-174). Chicago: University of Chicago Press, for the gifted and VAN TAsset.-BAsicA, J. (1981). Acomprehensive model of career education 7. 325-331. talented. Journal of Career Education, Study of VAN 'TAssEt.-BAsicA, J. (1983). Statewidereplication in Illinois of the Johns Hopkins Mathematically Precocious Youth. In C.P. Benbow & J. C. Stanley (Eds.).Academic preconty: Aspect; of its developnwnt(pp.179- 191).Baltimore. MD: Johns Hopkins Univer- sity Press. (Ed.), VAN TASSEL-BASICA, J. (1985). Appropriatecurriculum for the gifted. In.!. Feldhusen Toward excellence in .gifted education (pp.45-68). Denver, CO: Love. Mincemccgazoie. VAucatAN, D. (1984). Twyla Tharp;Launching a new American classicism 58(5), 54-58. VERNON, P. E. (1979). intelhgence, heredity, and environment.San Francisco; Freeman. VERNON. P., ADAMSON, G., & VERNON, D. (1977).The psychology and education of gifted children. Boulder. CO; Westview Press. WAGNER. II., & ZIMMERMAN, B. (1986). Identificationand fostering of mathematically gifted students. Educational Studies in Mathrmatws, 17(3),243-259. WAu.AcE. A. (1986). The pnidigy: A biographyof Willtam fames sai,. Amertra'y greatest child prodigy. New York: Dutton. WALLACH. M. A.. & KOGAN, N. (1965). Mod" ofThinking in young children. New York: Holt. Rinehart k Winston. WALEAs, G. (1926) Thr art of thought. NewYork: Harcourt, Brace.

3 !.4 382 References

WALBERD, H., & FOWLER. W. (1987). Expenditure and size ef liciencies of public school districts. Educatitnial Researcher, 16(7), 5-13. WALDERG, H., RAstsER, S., & PARKER-um, J. (1979). Childhood and eminence. Journal of Cre- ative Behavior. 13(4), 225-231. WARD, L. F. (1906). Applird sociology. Boston: Ginn. WARD, W. C. (1975). Convergent and divergent measurement of creativity in children. Edura. &mai and Psychological Measurement, 35, 87-95. WATT, 1. (1957). The rise of the novel. London: Chatto & Windus. (Paperback edition published 1967 by the University of California Pres.$) WECHSLER, D. (1958). The measurement and appraisal of adult inielligence. Baltimore, MD: Williams & Wilkins. WECHSLER, D. (1974). Wechsler intelhgence scak for children-revised, New York: Psychological Corporation. WEctistin, D. (1981), Manual for the Wecfuler adult intelligence scale-revised. New York: Psycho- logical Corportation. WEER.% T. E. (1974). The stow speech development of a bngla child. Lexington, MA: Lexington Books. WEILER. D. (1978). The alpha children: California's brave new world for the gifted. Phi Delta Kappan, 60(3), 185-187. WEINER, N., & ROMNSON, S. (1986). Cognitive abilities, personality, and gender ditferences in math achievement of gifted adolescents. Gifted Child Quarterly, 30(2), 83-87, WEINSTEIN, J. B., & Bosxo, P. (1980). The relationship between creativity and androgyny when moderated by an intelligence threshold. 14fref Child Quarterly. 24(4), 162-166. Weiss, D. S., HAIER, K. J.. & KLATIM;, D. P. (1974). Personality characteristics of mathe- rnatkally precocious boys. In J. C. Stanley. D. P. Keating, & L. H. Fox, Mathematical talent: I ;newer, description, and del,elopmeni (pp. 126-139). Baltimore, MD: johrs Hopkins Univerity Press. WELCH, W., ANDERson, R., & HARRIS, L. (1982). The effect of schooling on mathematics achievement. American Eduratioruil Reseanh Journal. 19(1), 145-153. WELSH, G. S. (1949). A projective figure-preferenre test for dsagrunis of psychopathology. Unpublished doctoral dissertation, University of Minnesota, Minneapolis. Wets's, C. S. (1975). erzativity and ridelligenre: A personality approach. Chapel Hill, NC: Institute for Research in Social Science. Weeco-i, G. S., & BARRON, F (1963). Barron Welsh art Wilk. Palo Alto. CA: Consulting Psycholo. gists Press. Where are they now? (1969, April 28). Newsweek, p. 18. Wistertx., G. M. (Ed.). (1941). Art in American life and edwatom. The Fortieth Yearbookof the National Society for the Study of Education. Chicago: University of Chicago Puss. WHITE, M. A. (Ed.). (1966). Sc/wol disorder, intelhgerue, and wird doss. New York: Teachers College Press. WHITMORE. J. (1980). Gif redness, conflict, and undenohnlement Boston: Allyn & Bacon. WHITstORE, J. (1981), Gifted children with handicapping conditions: A new frontier. Excep- tional Children, 48(2), 106-113, WHITMORE, J. (1982). Recognizing and developing hidden giftedness. Elementary School Jour- nal. 82(3), 274-283. WHITMORE. J., & MAKER, C. J. (1985). Intellealial gittednesA in disabled pes.sons. RINkvjlk, MD: Aspen. WioncroN, J. F. KARNES, F. A., & CURRIE. B. B. (1985). Comparing ability and achievement in three academk areas for upper elementary gifted students. Journal fm the Education of the Gifted. 8(2), 149-154. WIENER, N. (1954). The human we of human beingi: CyhernetUA and WO (rev. ed.). New York: Avon. WItaaNToN, E. (1985). Sometimes a sturunk moment. Garden City, NY: Anchor Press/Doubleday. Wiewx, K. (1982). Differential socialization in the classroom: Implications for equal opportunity. In G. Spindler (Ed.), Doing the rihrwgraphy of sdwoling (pp. 268-309). New York: CBS Colkge. Wit.cox, K., & MORIARITY, P. (1977), Secooling and work: Social constraints cm equal educational opport unity. Social Problems, 24(2), 204-213.

391 Referrowes 383

WILKINS, W. L. (1936). High school achievement of accekrated pupils. School Review, 44,268- 274. Ama, F. (Ed.). (1970). Language and poverty.Chicago: Markham. WILLIAM. F. (1972). A WWI creativity program. Engkwood Cliffs, NJ: EducationalTechnology, WILLIAMS, R. (1976). Keywords: A vocabulary of culture and society. NewYork; Orchird University Press. Wu-Lincs. D. (1985). The specific needs of adults who are gifted. ROIPPT Review,8(1). 35-38. WiLurs, F. F., BEAwl, R. C., & CRIDER, D. M. (1973). Leveling of attitudes in masssociety: Rurality and traditional morality in America. Ruml Sociology. 38( I),36-45. WiLsorm, F. T. (1953). Some special ability test scores of gifted children. Journalof Genetic Psydwlogy, 82, 69-90. WILSON, R. N. (Ed.). (1964). The arts in society. Englewood Cliffs, NJ:Prentice-Hall. WINEEORDNER. D. (Ed.). (19148). Career exploration and planning program.Bloomington, IL: Meridian. WING. H. (1948). Tests of musical ability and appreciation. British Journal ofPsychology Mono- graph Supplement (No. 27). Wm:. H. (1954). Some applications of test results to educatnm in music.British Journal of Educational Psychology, 24, 161-170. WING, H. (1955). Musical aptitude and intelligence. Education Today, 5(1). Wigs/A1P, 1. :1981). The Soviet challenge: The new mathematics curriculumrequired of all students in the USSR is superior to that (If any other country. EducationalLeadership, 38(5). 358-360. Wirry, P. (1930). A study of one hundred gifted children. Lawrence. KS:Bureau of School Service and Research. Winn, P. (1958). Who are the gifted? In N. Henry (Ed.), Education of theGifted, The Fifty. seventh Yearbook of the National Sirciety for the Study of Education. Part 11 (pp. 41 - 63). Chicago: University of Chicago Press. Wiwi, & Gvot, J. (1981). Learning disabled and gifted: Success or failure?Journal for the Education of the Gqted, 4(3), 199-206. Wocf, M. H. (1981). Talent search and development inthe visual and performing arts. In Balancing the scak for the &advantaged gifted (pp.103- 115). Ventura, CA ;',.tional/State Leadership Training Institute on the Gifted and the Talented ant, 0:ine of the Ventura County Superintendent of Schools. W000ccx.x, R. (1973). Woodcock reading mum tests. Circle Pines. MN...%nerican Guidance Services. Wootw.ocit, R & jOHNSON, M. (1977). Woodcock-Johnson psychoedwatienalbattery. Boston: Teaching Resources. WOODDELL. G. (1984). Gifted children in general music. Moak EducatorsJournal, 711(5), 43-46, WOODMAN, R., & SHERWOOD. J. (1980). The role of team developmentin organitatiimal effectiveness: A critical review, Psychological Bulletin, p8(1). 166-186. Waitart, E. (1979). Class structure and income determination. New York:Academic Press. YSSELDYKE. J.. SC ALI:MI.1SE, B. (1982). Bias among professim als whoerroneously declare students eligible fcr special services. Journal of Experimental Education, 50(4),.223-228. INSELDYKE, J., ALGOZZINE. B., & MITCHELL, J. (1982). Special education teamdecision making: An analysis of current practice. Personnel and Guidance Journal, 60(5),308-313. ZAFIFRANN. R. T., P. COLANGELO, N. (1979). Counseling with gifted andtalented students. In N. Colangevi & R. T. Zaffrann (Eds.), New vinces in counseling thegifted (pp. 142-153). Dubuque, IA Kendall/Hunt. ZALE/NIK, A. (19144) ::harismatic and consensus leaders: A psychologicalcomparison. In W. E. Rosenbach & R. 1.. Taylor (Eds.), Contemporary issue, in leadership (pp.255-270). Boulder, CO; Westview Press. ZANDER. A,, & VAN EGMOND, E. ( 1058). Relationship of intelligence andsocial power to the interpersonal behavior of children. Journal of Mutational Pswhologi, 49(5),257-268.

3 2 Index

AUTHOR INDEX Bowen. j.. 69 Bowers, J , IOC 102 Abel. 14., 230 Bowles. S . 104, 118, 159. 298. 299 Abelson, P.. 86 Brady, I1978 264 Adams, J.. 165 Braga, J , 528, 555 Adkimon.i.. 96 Brandwem, P . 151 Adler. M..352 Braunschweig. P... 205. 206, 207 Albert. R., 256 Brekke. B., 231 Albright, H.. 209 Bocklin, B., 268, 280 Aiesantkr. K. 154 Br sklin. P., 2011. 200 Alexander, P.. 55 Boggs, R 116 Aleaander. P.. 334 Wiliam, W.. 182. 195. 201 Alexander. R , 144 linwkway, W . 254 Akmander, R., 201 Btonlenbrenner. F , 05 Algoanne, B., 60, 161, 287, 289 3.113 Brophy. _I271. 299. 399 , 151, 354 Almy. U.. 231 Broudy, H.. 189. 192. 195, 194 Alter. J., 204 Brown. A . 53 Alvino.3 . 114. 325 Brown, I). 319 Ames, L, 308 Brown. j, 60.150 Anastass A., 98. 99 Brran. H., 2147 Anderson, R.. 177. 178 Bryan. 1 . 287 Andrews, I' . 91 Brush. 1.159 Anobi. S.. 239 Bruner. J.. 169 Ansbashet, H.. M4 Bryant, B , 214 Ansbather. R .149 But hmann, 4, Anyou. J.. 23. 35. 67. 271. 272. 296 599 355 Buemung, U , 106. 107 Atends.1. 335 K.,-289. 331 Arendt. . 334. 341 Bats I.. I., 290 Arnheim, R.. 195, 1944. 21r2 Burks. B.. 256 59 Amnon, . 191 Burns.) . 126 27 Ames. 43 240 Kurt Mo, K , 195. 196 Bushell. D , 270, 271, 272 Astro, II., 250, 251. 111, 312 Busse. '1 116 Austm, A . 219 Butterfield.. 65 Avner, F,, 161 Ruth. R.. 69 Ilsnum, C , 519 Lohman.' . 65. 66 Bat kman. 741. 106 t 411aban, . 44 Baer. D., 270, 271. 272 tkilahan. R 10, 349 Band. 1-. 50, 155. 394 Callawas. W . 185 Bakissm. A . 12 Camara, K . 65 Bamberger..1 , 256 Campbell, 3 Sol Barbe. W, 157 Lai . 274. 2147. 509 Barbee. A , 136, 312 (14Troll.. 155, 161 Barker. R., 51$ Carter, K. 13414. 225. 229. 780. 71), 732 Barnette. L.. 85 l.arter, R 60, 61 BatTrit, F.., 214$ t.arrwright, C . 274 Barron, , 88. 91. 92. 91. 204, 2044 t:kownght. .34313. 274, 285 Bar-Tal. D , 2E9. 278 t".asserty. P . 511 Barthe. C.. 165 Castiglione, 1.. 335 Barylkk. U . 208 Calm. 11 , 2142. 293 Barium 1346, 1$7. 191 (..4641. A., 51 Bass. 328 Cana, R . 5) Bass. B , 328 j., 225, 226 Bawls% A , 126, 150 tbno. 195 Bayley.. N., 42 Chank4. N . OH Heider. R.. 515 Chauvin, J .131 Benbow. C., 174. 176. 2414. 252 51. 811 Chrtelat, F. 327. 350 Bennett, W.. 73. 340 Christenwn. P , 101 Berg. 1.. 139 Clarnm, H ,155, 309 Berliner, U. 29, 31, 110 Clark, 13 . 96, 164, 165. 217. 233. 335. 3811. 540 Bei-WIWI. W.. 169 Clark. 6 , 183, 197, 1911. 199, 200 Bernstein, B , 503, 304 t lark. P., 99. 100 Best. D.. ZOI Clarke. A 83 54 Bierly. K.. 103. 333 Clarke. E 11916119610, 119 BUtet. A.. 39 Cobb, 4, 150. 152 Birch. J.. fa. 156. 261 Cohn, S, 311, 312 Birnbaum. 4.. 314. 315. 117 Colangelo. N.. 233. 235, 267 Blachly. P.. 240 Cnirman,, 25. 55, 36, 37, 55. 317 152, 237. 279, 272 73, Blau. P., /NS 296 Blaubergs, M., 509 Cannel. _f. 255. 279 Bloom, 8.. 67, 120, 154. 170, 190, 254. 255, 279, 272, 348 Cnplev.F., 333. 3311 Boat), P , 03 Cornell. U. 64. 235, 236. 237 Boehm. L.. 233 CAYunts, 6. 296, 318 Boles, D., 175 C.ox, C 119-20, 246 Bolt, 3., 137. 140-44 Cox,, 16. 17. 64, 73. 76. 77, 165, 167, 176, PS, 291, Berta. . 69, 96. 272 256. 316, 557 Basins B., 2147 (JAW. 1414 Boston, B.. 16 CI'IrMLII, I-, 72 270. 341. 343

384 3 I Crider. 315 Forntss. S. 288 CtteSS, 11.263 Foster. W.. 291. 245 Grouter. H., 65 Inuits,. I. 72 Usurder. 1.. 62, 149, ISO, 152 Fos. 1... 99. 153. 154. 157. 226, 230. 251- 52. 311. 312 C:sikszenunthalsi, 54. 19$, 200 Frauenheim, J288 Cuban. L.270 Freedman. D.. 42 Cumtip. S.. 316 Freedman. N.. 42 Cummins, J.. 161 trreltill, M., 318-19 Cunningham, A , 137. 225 Freeman. j . 155. 253. 2114 Currie. 11.. 150 French, j.. 270 Cantu. M. 161 Friedburg. S.. 168. 178 Cutler. R.. 291 Friend. R.. 189 irsr, D.. 210 Fung. 14 . 236 Dasrv.3.. 101 Daniel. S. Ili, 21.11 Dansin., 214 Gaddes, W . 508 Dattriu, S., 334 tiage, N . 29, 31. 310 Dandron. H.. 223 Caaglwr.. 9. 35. 62. 68. 73, 114, 131,199.150.152. Davis. , 131 164. 222. 233. 237. 278. 286, 135,3311 Davis, G., 93. 267 CtalIuri, N., 234. 290 Davis. H . 254. 279 tialton. E.. 4. 5. SO. 18, 99, 81, 87, 88 , 90. 93.189,214, Dav6.1. 120 219 Davil,7 , 169 Gardner. H., 3, 14. MI. 91. 162. 189. 11414.200,209.244. Davls. S301 502 De Wool.A . 264 Gaug. 54. 162. 194-5. 166. 167 De Leon, J.. 306 Cieras.c. W . 351 Debate. j.. 231 Grterlsicti. 90, 91, 94. 99. 101. 102. 103.1114,184,198. Demos. G.. 337 Denbo's. C.. 64 (ihiselli. E., 129 Denham, S.154, 250. 231. 252 Ginsberg, A.. 65 Denman. B.. 204, 209 C.intis, . 194. 1 IN, 159. 296. 299 Ikutith. C. 42 Ginsberg, E . 133, 115 Deutsch. G. 108 Giroux, H.. 101. 353 Deems. j.. 945 t;orns. K. 304 DeVoung. A . 313. 349 (Hass, G . 252 D'Iletuk. A . 235. 233 272 Dnnurovsks. L, 231 Goensel. M.. 118, 121) Dishes. W . 240 Ikaeorrl. V., 1114. 120 1)o8son. R., 116. 126 (;04d. M.. 139 Duhahanskv, I, 42 Goldbers, M , 15. 298, 278, 279 Dolan. 163 Good. '1. 271. 299, 109, 555. 354 IkairLeg, D.. 274 Goodwm, W , Ikm1W, C . 267 Cond-in, F... 189. 190, 195 Draper. B, 219 Gordon, W . 104, 167.14314 Drew.. F. .451 Cola. K.. 91, 92 DntwILl.. 102 Gtutkl,S.13, 29, 34. 35. 38. 39, 42.4$, 154. 301, 504 Duke, R., 270 Gouldner. A . 116 Duncan, A , 61 Gowan. j.. 1. 103. 337 Duncan, 1, 206 Gray. R . 221 Duman, 0 , 136 Green W.. 284 Duntum, P., 193. 1914 (,reenfield. P . 159 Dunn. L. 162. 176 t;liffin. 1. . 134 Dunn, L. 286,287 1:offith. I) , 99. It84 1)upuo. M., 160 Gum. A . 62, 246 Durden. W.. 166. DST. 222 Cloth, N . 65 1Furkm, U. 66. 157, 165, 225 26 Grubb..1 . 277 Duo', W. 165 Guilford, J5, 50. ME 82. 97. VII. I57 135. 221 Dykes. M , 2911 Ciump. P..113 43suhrie, 11. 288 Fiks.J.. 175 Edge. 0. 198, 178 Edmiston. K . 61 llagprti. h. 231. 289 Edmonch, R . 399 Harer, R., 154. 299, 230 R . 152 Haller. F . 501 Lpstem. C 137 Hamm& I).. 162 'Aunts, C.. 213 Harness. B., 309 1,rixh. V. 165. 239 Hants. M . 177, 1714 tylantk. H., 86. 92 Hartman, K, 49, 93. 158 F.V4C114k,S.. 92 Flatkell. S , 2/414 liathawin. S.. 92 Hawk. IS . 518,- )9 Frain, 1, 160, 164, '222 Harden. S . 262 Featherman. D 'laves. C.. 63 Ferman. D.. 157. 225 Heath. S. 1412 Feldhusen. J., 214 Ilreren, V . 169 Feldman. D.. 68, 102. 1149. 244. 255 Herd. 84. 170 Feurrstrin. R 42, 119 Ifehon, R 511. 512 Fitcrfla,1 Henson. I , 528 Fen. 0. 189 flerrnstetn, K. Fine. B., 268. 279 . 177 Fink. A.. 291 Henry... 114 Fink. M , 2714 irrsh, . 169 K . 11 Hrthenngton, F, 65 Fisher. R 253 /lichen. E . 224, 225 Havel). J . 227 ltddreth.0 , 40 Fleming. f. 131 Hocrvar. 83, 84. 99, 1181. 164. 327 Fontesn, M 206 Hoffman. 54., 42 Lord. W , 168 Huktadter. R . 70. 116, 117. 104, 126, 154 Forman. S . 99 J . 152, 249

3 4 385 Manmade*. L. 194. 253 laitertatw. 162 tballingworth. L.. 11. 63. 72. 130. 152. 137. 164. 170, 217, LaBarge. D.. 223 221. 223. 236, 238. 245. 3214 Labos. W.. 159. 298. 303.103. 304 Horn.). 33 LaChapelk. J..113. 84 Homer. V., 191, 192 Lane, V,. 209 Huns fray, B.. 162 Lang. D , 51 Howard. L, 304 Langsray. L . 218 Hewn, R., 281 Lanier, V.. 201 Hanky, A., 16. 73. 76, 104. 112, 116. 118, 126. 1411, Lawn. 5.. 221 151. 176, 177. 1714. 191. 123, 529 , 333. 333, 316. Latruck. K. 216 345 Umiak, E .110 Murky. C , 16. 302, I. 333. 334, 336 ler. L. $05 W , 152 Lehman. F , 2901 Huff, V,. 103 Lehman. N., 235.105 Hume, 14,, 286 lalatimm. 1... 287 Hunt. E... SS, 225 Liontief. W . 71. 116 Hunt, ).. 30 mguld. A., 225 Hutthans, C . 338. $49 ewuntin. R.. 10. 32. 36, 42. 3411. *12 Huxley, A . 103 j.. 60 1.esser, t;56, 151. 34M Letris. 9 , 274 loklm. C.. 311. 113 Lan. H.. 104. 153 ackum, N., 60. 61, 160 Lipset. S116. 126 Lines. P.. 271 ackton. P. 94. 99. 101. It42, 103, 104 Litarngm. W.. 127 atom,. P., 23 Tobin. W.. 159, 160 acult. E.. 205, 2401 j , 53 alum P.. 236, 237. 23$. 32$. 334 latinhrosta, C . t17.14ti, 90, 233 muck.] . 176 Loughhn. 1.. 6,3 muck. S.. 176 latpair. P . 117 rnika, C . 50, 54. 104, 1211. 135, 136. 115, 119. 140. 270, 296, 341 latiornfrid. V . 152. 1115, 201 rnkins, M.. 245. 304-3 Luc nu. 1. , 215 enkins-Frodinan. 9 . 114 1 ton. If 114 mango, H.. 127 morn, A.. 1M, 34. 15. 134, 305 amen. D , 25/4 - 59 Maiciabv, 1.333.338 mien. 1.. 139 Macintosh. P. 2446, 2/47 *num. 9 . 269 MacKinnon, I). Kfi. 90. 91, 23/4 ohnion. /1 , 274 Mac Imsh, R . 244. 259 uhmon, 1 . 231 Madato,1:101 ulonno. M . 162 Maddo, S , 93. 214 utursno, P . 200 Madduri, C,. 32$ mes. L . 207 Maelu. . 61. 281-2, 253, 254 Mame,. P. 175 Makel. t.. 12. 251, 315 Kahnowskt. A.. 261 Mann. P. 29M Kalleherg. A , 134 Mansfield, R , 11N Kamerman, S. 65 Mari use, 13 , 83. MO. 93. 193. 341 Kamm. L. 30. $1. 32. 31 , 36, 42 Marionhankt, K. 101. 102, 3116 Kamphaus. R., 51. 73 Mark. 54. 191 Kamm. 11.. 269 Markmodt. 1.. 162. 176 Kantm-. M . 272 Marland. S . 6, 7, 16, 130. 134. 135. 270 324 Karnes. F., 131. 150 Mammon, R . 62. 75. 150. 344. 305 Karnes. M., If.3 Manmod, K , 237. 23M Katthadnourian. If137. 140 44 Mariano. R , 338, 349 LW,M., 270. 140, 342 Maslow. A , 89 Kauffman, J., 219. 240. 291 Mather. N . 267 Kaufman, A. 61. 75 MilYbrft,. M 313 Kaufman. N.. 51 Mc Ashur. M., 33 Katak. K., 287 McCall, M.. 132 Kratinit. II. 81. 153. 154, 229, 248. 249. 251 , 914. 2/44) Kelly, M.. 247 %Connell, t254 KeIlt. T., 290 McCracken. R.. 165 Kerah. M., 177 McCurn, J . 973 Mumma.) . 292 McDonnel, K. 325 Kimball, B . 265, 277 Mtim,rr. M 221 King. M., 291 P. 205 D . 8. 217 %Kamm.. 99 Kirk. B . 277 Mt Knight, 1: . 177 Kok, S.. 9. 256 McNemar. Q.99. 101, 102, 117, 155 Kirp, D.. 24 Mrdmck. M.. 914. 11M. 101 Kirswin. 1.. 203 3,. 96. 98, DM, 101 Knapp, M .14. 217 Merhl. P. 92 Klausm-rin, II . 61, 214 Mehrrns, W , 155 Mem, R.. 83 MrIlmger. A , 231 Knestkr. A . 83 Mercer. j , 40, 299 Kogan, N 85, 94, 96. 98, 99, too, 101 Mercy, J., 316 Kubn.M.. 299,100 Merrill, M . 222 Kontos. S.. 925, 229. 210. '231. 232 Mrs's% 1.. 1444 Konen. D.. 130 Mrssr. S.. 14$ Korrenik. P. 195 Mrstre. J., 351 Kraus, 12 , 93, WM, 708 Mrtcalfr, R , 999 Krutetskit, V..I I, 130. 152, 154, 155, 165, Is9. 17075, Memo.) 104, 1411 177. 179 M,tJueLW.,251 Kubarrn, T.. 60. 96. 2'72 Maks, P. 308. 3454 Kukk. C.. 328. 334 Miller. C.. 42. 169 Kuhk, j.. 32$. 314 Mkt. R., 233 Kurtzman, K., 312 Miller. S.. 187. 194 Kurkskv. W-. 314 C116, 174,125. 126, 316 Kwalwasser. I. 1$9, 190. 192. 194 Milnr. A.. 65 Kyle. P.. 136 Mame,M.. 157

386 395 Reisman. . 177 laintiket, V.. 42 4. 6, 49. 73, 95. 1514. 236, 113, 519 *Oen, 163 Renault*. J., Mio-ben,.. 278 Resintk, 1..164 ., 8. 163 RAMOk R.. 304 1411,thefl. Reynolds, (1.. 51, 254, 239 ?V.J... 189 J.. 260 Rite. W., 148 Moinago, A.. 42 %then. S. 325 Moore. t:230 Risismond. K . 254 kkatganbesset. S . 122 Riddrl. M.. 314. 317 %nun. S., 95. 267 35, 296. 343 R..60 Z=1.11.. Ratio. L. 274 144011311011, A.. 161 328. 354 Ma:nom E.. 231 Robinson, N., 60. 61, 160, 256, 237, '2314. Morse. W , 291 Robinson. S . 230 Mosher& L. 161 Rubinson. S.. 319 Motel/et. F., 33, 296 &whom. G.. 240 Moynihan, 1).. 35. 296 Rae. A., 91. 134. 137-38, 152. 154 Minnow/. H., 35 W ardell. W.. 614.61. 11:11. 210. 217. 234, 23S 187. INN, 193 %wiser. A , 279 Miarw Rogers. C.. 83.148 Myers. hi Rose. L.. 104, 183 Myers, J.. 103, 117 Ruse. R . 103 Rose. S . 30, 35, 56. 42, 24:04, 299. 5111 Mskkbusi. H . 274, 287 Rosenbaum, J., 47 Rosenthal, A .435 Nattanial, P.. 305, 316 Rosenwald, P..207 NIwmow.i. 151 Ross, .165 Naussipsky. N . 29 Rods.114 , 2714 Rothschdd, J. 122 Nthon. C. 290 Ruhsion. P.'. 61 Nelson.Piens. 230 Rumberger. R 435, 136. 158, 14(4 Newel. A.. 3311 Ruston. M., 256 Norton, Lt.. 2311 Rran. W., 36 Nye. R., 191. 192 Riwr, 1: , 2014. 204 Nse, V.. 191.192

()ain. J.. 53. 47. 178. 296 Sahmw. R., 199 ()den, M.. 34.544. t42, 64. 65, 67, 1147,1644. 1744.222210. Siwta.J. 325 245. 259, 2614, 3214 Samsida, R., 34 Samuels. J . 225 Cklin. A..119 Oiler., )56 Sado. 1. 221 Olson: 31 . 308 Sauter. J.. 31, 73 Orr). K. 191 Satinden, K. 199 Oinitud. J.. 166, 229,2311. 211, 212 Sawyer. . 117 Orion, S., 287 Starr. S. 33. 43. 44 ()shorn. A., 10. 85. Sti, 91, 104 haefei. U.914. 99 Schaefer, 1. 127 ()damn,W , 3314 Scheurle.1 , 221 Osten, , 3419 Stbieher. I.. 528 Shur:damn. 240 Scholwinski. F.. 234, 259 Paffentuager, K , 240 Salsonberx. S.. 221. 254 Par. 244 4whuhert.-1) . 99, 101 Paussu. M..159. 2112 Schuler.R., 1103 Parkersnn. J 114 Schlosser, 1.. 310 Parnts.S. 103. 1114 . 127 Stott. 34. 132 Partnagr. Som. 214 Pantos. A.. 35, 28$, 278 279 Scripture. F... 269 Pasternak. S. 252 Seagoe. N . '215, 216 Paatirre. N,. Peshods. J . 283 Seam P . 13/4, 112 WI, ¶44 , Seashore. C. 191. 194 Pearlman. C.. 83. 44. 2511, 259 Pefriato, t:61. 156. 521 Seeks. K . Prises. N , 252 Segal. S.. 118 Pendarsis. I, 160 Sas. L . 513 C .225 Sett!, ti , 288 'Kt 93. 101. 1113,104,I xi 184 Sessions. R., 11411. 191. 195 Perlin*. 1). 45, 84, sm. Seictim. P . 61. 510 Perkins,S. 24/4 Perrr. 1. , 116 Shade. B . 505 Sham:. C., 251 Peirold.R .1t47./WI Pfeiffer. S , 165 Shaw, NC, 275.'177 PfirT. 1.. 255 She:. J .111: Sherwood, J.29 Ptid1ps. S , 509 190 Phipps. P.. 509 Shrtler. 11, Pbunthas, J . 515 Shields. 1) . 32 Ilmskur. P. 1tO Much. N. 104 Paean). K . 2149 Stills. I. ,1311 Plowman, P . 517 Stunn.161.. 287 Sbonhur ft, S , 221 Poulamras. N . 2914 192. 193 194 244, 255 Shuirr,R..184. 188. 189, 190..91 Presses. S . 508 Prichard.A. 218 Silver, 3., 226. 279, 2sfi, s(4s User. R.. 349 PongW. 34 , Silverman. K. '201 Puri. P., 2744 !Onion, II . 338 Nisi, N.,272 Simonton. D.. 91. 112. 111. 101. '244 214$ Sok, D.. 105. 553 Rand, Y.. 42 167, 29S Rankin, 1.. 283 Suer. 1., 104, Raph. J.. 268,27S, 274 Skeels. It,53 Skodak. N . 33 R.ashn. S 114 R , 554 Read H., 71.85.196 Redden, N . 281 Smith. 1.. 49 Smith. L. 272 Reid. 1)162

387 Smith. IL, 161 Van Egmand, F.. 127. 131 Smith, S., 308, 310 Van laissel-Baska. J., 133. 138. 223. 233. 333, 338 Smith. S.. 262-3, 286. 2147 Vaughan. 9., 203 Samba. L. 221, Vernon. P.. 148 Strugsr. D., 3S Stiatio.:., 55, 312 Sone!, W.. POI, 203. 2 09 Wagner, H , 253 Si:umiak, L. 190, 234, 21! 256..8 Walberg, H.. 114 Spearman. C., 49, 152. 1: ' Wallate, A.. 63. 243 Srpoia. L. 262 Wallach. M., 85, 94, 96, 914, 99, 100. 101 Springer. S.. 108 WaRss. a.. 06 Stalker. M., 196. 197, 1914 Ward, L., 119 Stanwhinkt, C., 209 Ward, M.. 274 Stanley,, 60. 73. 7' T7. 153, 134. 163, 174. 17X. 248 Ward, W.. 99 250. 253. 311. 333, 314. 339, 334 Watt. 1., 70 Stanovhch. K.. 157, 224. 225. 226 Weathers, D., 304 Stedman, L. 57. 73 Wethsler. D., 50. 97 Wrist. R., 2143 Weeks, 161 Stern. H.. 161 Weiler. 9, 16. 23. 714. 3214. 313 Stem, IV.. 49 Welles. J.. 114 Stembers, R.. 4. 52. 53, 34. 36, 162, 333 Weinberg. R.., 33, 41, 43 Storkard, J , 61, 313 Werner, N , 86. 136 Stodgd1, R T. 109 Wemer, N.. 250 Stadohky. S.. 36 Wonsserin. J.. 81 Southey. J.. 814 Weiw 1),. 249 Strang, R., 157. 160 Wekh, W., 177. 1744 Strauss. A., 2147 Wrhh, WS, 96. 97. 99, 100 Smnget. 1.. 272 Wharum, 1, 130 Strafteryes. 1:., 262 Wheatley. 9, 177 Suchman, L. 152 Whipple. G.. 197 Scilhvan. E., 230. 231 White. A. 49 Sularia, S.. 287 WIntmnre. I. 157. 226. '270, 274. 275- 76, 281 82, 2141. Sward. K., 127. 254. '253 264.288. 09 3w/wing, R., 335 Wrinstack. H.. '234 Syphrn, 11. 1314 White, 54 , 790 Wikus, K , 23. 15, 1(4, 27). 296, Mat 111111 Wilkins, W. 163 lab/. H.. 349 WiRerman, 1. . 31 lannenbaum. A , 9. 35. 61, 73, 85. MI, 117. 167. 237. Williams. F. 33 235 Wdbams. F. 167 T aa, 8,247 . 231 Taylor, C. 112, 152 Wdliams. IR., 297 wkw, J , 195. 2114 Writings, 11. 139, 143 Inlot. 144 , 288 Wdbts. F . 315 lafendler, C.. 236 Wthon. 130 51 rerman. 1... 3. 3. 54. 39. 40. 56, 1i2. 67.1114. 149. 1541. 113. Wilson. R.. 204 157. 1514, 160. 1137, 215, '217. 233,235.216.'240. Wincup. 1, 170 245. 2514- 59, 21/8. 299. 3142, 3214 Winelardner. U. 155 Tharnrbkr. R.. 30. MO. 102. 116 Wing, 14 , 191. 191. 144 thrashes. Y , 127 Wahi8., 122 Thum.% 1... 71, 135 Waty. P . 5, 62 humane, L.. 50 Wolf, M . 209 .4 dwe1l. K. 35. 117. 133. 1101. 194. 197.2:48.145 WrOt. ).. 2104 lunar. 1.. '24 %Void. J.. 61 Tndall. W . 283, 319 Woodror k. R . 162, 1713 lulkfum, N 306. 307 Wsifiddell, , 190 Tonelson. S , 2141 Wnght, II.. 113. 2911, 299 Tunemall, 5 , 306 Coirame. F . SI. 85. 94. 96, 97, 99. 100,14.4),1142,103, 104. 117. 129. 112. 167, 184, 290 S'swldike. J. tat. 013. 287. '289. 325 Fraven. K.. 177 Trense. R.. 165 Fyark, 1,1 . 113, 541, 342,149 talltan, R . '233 '1 der,F. 157 talernik, A , 132 Zander. A , 127. 131 Zimmerman. 0.. '253 V&A A , 267 Zimmerman. F ,INS, 197, 198. 199, 200

SUBJECT INDEX And unfair &stimulation, 294-320 mad am, 198 Absolute pitch. 190-91 At:hien-mem sews. 7. 47, 58. 39. 60. 61. 62, 74-76, 99, Academic 102. 148. 149, 153. 156, 163. 199. 249. 1167. 273. stably. b. 7. 13, 61. I11 2713. 277. 266, 2/94, 291. 293. 315.123.1125-26. perform/me. 53 333 Wein. 18. 24, 146-79 Adaptive behavior. 49. 34- 58. 714 &relented. 76, 157, 164-66. '218. sr& 330 .44pebw Chentint, 249. 250 Acceleration, 72, 73, 78, 164, 177, 178, 332-37. 540, 349. Adler. Alfred, 89, 103 335 Adeleseew Sonny, 23. 272 of concept derrlopment. 250 Advanced Placement, 311 and dance instruction. 208 Aitport-Versol-Lesdrier Medi ri 96, 91. 249, '251 and music instruction, 104 Alpha Children: Califinnia's Brave Nea World far the and virual ans Instruction. 201 Gifted, 21 Accelerative. 76. 249, 345 A11117111215 &kiwi Board formai, The. 334 Achievement. 28. 30. 34, 40, 51. 52, 62, 63. IA. 66. 74, 96. Aptitude testa. 7, 235, 2414. 2145. 524. 326. 527 101, 216, 230..35, 236. 244. 245, '230, 253 scholastic. 55, 61, 76, 714 and creativity tests. 102. 149, 151, 155, 161. 162. 163. music. 193 177 Apphed Sorwim, 119 inhibited. 265-93 Anmintetarat Pradires. 24)3 musk. 192 Annex, S. 16. 23 precocious reading, 223- 26 achwyetnent, 40

388 397 precocity. 197 Dance. 185. 201-9 talent, 182-211, 321. 327 Dermition 01 giftedness Amara is theMabee. 93 and identification of giftedness. 323. 325. 327 Movement. 323, 325-27 instrumemal. 13-10 Athletic ability. 4. 127 pedagogical. 5-8 AthIesk premain. 260 -61 theorem-al. 3-5 Delinquency. 233, 239. 290 Desiation fromthe mean. 4.bee atm Suridard cleriumeno Barron Wfish MI Seale. 97. 99 Dewey...101m. 341 Behavior checklists. 40, 95. 96, 315 Mhistomous ability thrones. 48, 51-52, 78 Behavioral theories, 48. 49 Disadvantaged. 35, 40. 73. 207.210. 219. '271, 311. 3 IN Binet, Mired. 15. 314, 444 Discrimim asn. 295. 295-320 Blacks, 210. 258. 311 Divergent, 10.141-84. 94-95, 103, 112, 129. 151. 1144. discrimininson. 61, 290. 296, 299. 310 270. 271. 289. 315. 337 family suppom. 65 Doman, Glenn. 218 ct,abtfiereners IQ score, 94-36, tin Drama, 185. 244 y of IQ. 41 Dual aneodance. 76. 336 identification as gifted. 28 Durkitie studies of earh readers 2214 25 mean ig score. 21 and subeestive identflicaminuguruntrIa*, 20 verbal abibly, 302-7 F..nly entry. 336 Brain capatity.441 FArly tra4rn. 17, 157, 219, 222.26. 2144 Brain function, 308 t.erh Readers, 223 Bram meant', 314 Falls reading. 66, 162. 177, 222-26, 240. 241 %rota, Ilauk 314 Einvoirea rod * cubty Ell mann,9 Eduiation of MI Handicapped Children's Au nt 1975 (ire Pail's law 94- 142) califisnuaArammerist Test1419 fedi-tic imagery. 261 Cahlanua Prnewalin Index, 93 fine. 1114, 116, 316 (aahlarsta rothdagital Inienturi, 249 Elitism, 22. 25. 28 Coneenue andIstrarinashne Among College Sharne..140 farst. 320 Cavern. 107-145 Eminent. 80, 81, 87. 88, 114-18, 107-145, 254 Career imerreb. 251.252 Emotional demlopmem. 233-240. 334 Ceiling. (we 'lestciting) Emotional problems. 290-443 Characteristics of giftedness. 1017, 119,323. 327, !t33 Emotional stability. 234-35 areistic ability. 194-99 Emotion" disturbed. 274 bilatk *chimera, 303, 305 Enghah MniofSmarr .147 careerists and imam tuals, 140 44 Intschment. 3. 72. 77, 70. 164-65. 167. 1714, 194. 201. chess prodigies, 264 208. 236, 333, 337-40, 354, 335 early reader*, 2214 Environment. 27-43, 51, 54. 36, 59, 61. 119. 1149, 194. high.1(2 ehddren. 178 217-111. 221. 225, 311. 327, 3214, 329 leaders, 124-131 hivtraimental. 23. 30, 41. 42. 2'20. 249, 2140, 309 mathematical shifts/. 170-75. 178. 179 EqualAUSS. 23. 74 mathematicalls pima sous students. 2414- 52 I qualify 4 FitistatianalOppormion, 36. 299 niumal prodigies. 254 58 kiror in testing, 20. 21 royal families and immunities. 315-- 111 tapiessne lazunfe bee ?inductive language) rural gifted, 314- 15 Forma Poi amatory. 249 rural schools, 517- 18 kali-reek Personal" Queinaurimo, 92 underat !nevem, 270. 274-79, 291 verbally apt, 1744 Cheeklists, 4w liehavan thetklists) Child prodigies. 243-264 Families ill gifted Chddin Wish Abair 180 IQ.223 early rea*rs. 223.224-25 Cognniye. 98. 99. 100. 105, 154. 183. 1146. 195. 220, 221. eminent individush. 1211 240, 247, 2414, 250, 2417, 271. 507 high-IQ children. 63-67 complexity and art. 198 literary prodigies. 258- 59 developnrem. 153. 225. 256. 291 musical prodigies. 256-57 press, 164 in rural communites, 115 -19 processes. 4. 441. 50. 52-53, 155 56 and lotroetnnomicsutus, 297 99 sigh, 131. 132, 203 underachieyen, 291, stiles, 306 world-slainathletes, 260-61. 263 Caburn. Zerah.263 federal defininon el giftedness. 5 7 Carp En:114W Examination amidttl.FIlt,249 fluid mtelftgenie. 51-52 Concept deselopment. 227-.32. 2/44. 351-53 foreign language aptitude. 160-61 ConceptMaven Text (4M.11.67. 99, 153 fOreign language learning. 1.56. 179..344 Content.onented, 337. 339 heual, Sigmund. 147, 1414 (..41;19110LIT WINILM. 329 - Convergent. 141-84. 944. 100 Croturt 4 Eausnarr. 120 g (general tual f.0 urn. 35, 49, 152. 047 Creatise. 182. 184. 1146. 204, 214. 21N. 231. 234, 259. 2$1. 1; P.A (see Grades) 297. 325. 327 Gage. Nathaniel, 555 4:reatme personalny. 07-93 Eranto, 48. 55, 215-16 Creative thinking Ada,. 73. 135. 323. 557 Gardner. floowd. 50 and types of giftedness. 6, 7 Gender (ire Sea) Creammy, 3. 4. 10. 13. 20. 444. 55,141, 120 Genius, 54.147. 113.214. 239. 244 arts talent, academics and 1142 --86 Gramni Rnidnier, 247 tens, 40, 164. 183-84. 199, 21). 251, 326-27 tidied and talented Chiklren's Ac1.1 1978, b. 130 31 rnital thinkmg. 24, 79, 34,40 Gifted-44w Shams. Thr, 311419 t ntenon.referenced tests. 162 Gtfird Mid Gine vp. Thr, 240 ( rims-class placement, 76 Glird adder*, 8 Crystallized intelligence. 51.-52 "Gifted Education arid theSMI r of Elitism.- 23 (uliurr.fair assessment, 74 (4ted Croup at Abair. The,67 f abler-Fat, IntrthgrateTeni. 51 i;ortionMaw ArW.Priafik.193 Curriculum. 28. 61. 72, 77,78. 155. 164. 34414. 1714-77. Grades, 47, 50. 60-61, 149. 156. 246, 267, 275, 276. 291. 316. 333. 335 337. 3314. 339, 3414, 350. 555 309. 310, 324. 334, 315 academics and the arts. 54-47 Grade.skippmg, 72. 77. 178. 335, 336 achievemem and. 270. 2714. 281, 281 4;roduatrRivardExam:maws (GRE). 62 arts. IRS. 210 Greet*Taal Cakulaksi, The, 262-65 donee, 207. 2014 Grmip/oventortforEning latemrsts(GIFFI). 95 hidden. 300 (owl*Inienttos for Finding Talent (GIFT,I.95 musx . 194 Guilford..f.,

389 Handkapped, 203-274. 328, 330. 331. 942 Jung, 9, (i .0414 Handiuipped gifted, 260-293 juvernha. 119. 2514 Handicapping condition. 267, 293 Hearing impairments, 2I49-M Heredetarian. 18, 19. 2$. 55 . 36, 39, 42. 41. 30. 189. 220 Aatipwon Antonini/ Raney 1X-AM:). 52. 7t, ffiredttart Gown's. 4. 49.147, 1/9 Kintets1M. Vadim, 170, 75. 251 Here& 27-43. . 119 Kluteakii's study. 170-73 Ir, 33. 41-42. 47 Hegberiorder thinking, 333. 335. 3104, 348 H1g14-I92 children. (ire also IQ) 16. 17. 149 -51. 153. 155, LanLuczpalquisnicm. 220-21 156. 159, 162. 176, 178, 212, 41. 245-47. 259, 328. tr 346 :Auk eminence and yeti-cis, 107-143 223. 245-47. 328 fives of giftedness, 6, 7, 8. 16, 20, 46, 323. 328, 341 theykgifted,students, 304-5 (we Learnmg disabled. Learning disabihnes) Hispank, 20, 214, 36, 65, 210, 296, 300 learning disahlhOn. 274. 2141. 286-911 HO& MA Vega:sows] Perferetwe inprOon, 249. 252 training &sables!. 271. 2$1. MM. 309. 319 feta. 11. 72, 170. 245-46. 341 Learning Mr, 49, 53. 62-13. 2914, 529 . 270 feu Than Wants Can Sul, $444 Himuch Van We Boost IQ and Scholasin Athwi. "Lightning" calculators. 155 mem?. 30 Lincoln School, 31$ Rea, iya Truth Faso Rain to Read. 218 listening comprehension. 156 57 Humanistic psychology. $9. Literacy. 68- 70. 73. 74 Adana, 195, 305 musical. INS Identsfic anon. 4, 164. 179, 323-27, 133 literary preen: its, 258- 61 ailing ukni, 210 total nonnin. Ana lakm. 163444. 166 !Altai norms. I4. 302 dance talent. 206-7 !Anti% it) iontr411. 269, 27'1 earls traders. 224 emolionall, disturbed, 291 Oh' mathemanial difficulties 309 Mai Arthut Fellow's, 144 math ability. 175-76 5fits, 4 Grwus. The, 87 minoria gifted. 1111-2 Mathematical abihty, 4, 148. 149, 151, 152. 153, 154. musttal prodigies, 254 155, 156. 167-714. 247, 344 musical talent. 193 Makougutrual tapenencr, Int, 169 rural gifted. 314- 15 Mathnweihrai idrai. 1149 visual arts Lakin, 199 Malheinatical pin m . 247-53 Inequality. 34-.35 Maihemancal skills, 1614 -70 Insuuctional formats, 347-50 Mathematical talent, 11 Instructional techniques, 347-30 moskmatwi the Art q Rniwn, 169 Intellect. 46. 68-72. 88. 114 IN, 169. 250 Mental disturbance. 239-40 Intellectually gifted. 46. 614. 71. 73-78 Metatognitise. 53. 338 39 Intelhgeme. 3, 2$, 34, 1143. 189. 144. 1414. 214. 217 1$, Miniwits groups. 33, 34. 411 41, hi, fiti. 74. 75. 2 tti. '272. 239. 241. 249. 273, 281, 2142. 284. 288. 291. Pm 301. 20 adult productivity And. /40-81. 87. 95. ltr.) ifnusecula Alufrrlikasti Pri.unudsec finentyn. 92. 93 (mamma, 44-76 Mowecoun 4 Man. Or. 29 definition. 47-4$ WI hell, Rig hard, 344 group differences. 55. 38. 1)2 7, 323 Mulupk-fastoi theories, 414. 50-51, 78 normal and abnormal. 46-54 Musical abilms. 4. 50, 164- $5 quotient. 49 Musical prrs nt 184. 251- 5$ tankuig. 46-47, 53 Mutual Went, 1$9. 93 theories, 44, 53 Intelligence tens. 15, 34. 35, 19. 40, 42, 47. 56 19,?, 49. 47. 50. 52. 57. 50, 81, ri5. 96. PA). 264. 3)2. 114. ..ccsilan at Ruk, .4. 9 315. 31$. 324. 327, 390. 131. 332 Nanonal Merit Si holarthip }.1tant, 249 academic talent and, 148 -57, 161 Nature. 28, 29, 189, 200,130 Achievement test scorn And. 39- 60 NomserbAI tests. 34, 248 Adaptive behavior and, 54, 56 Nol in- referenced tests. 149. children with high wares, 72-78. 129. 212 241 Nurture. 214. 29. 105, 189. 195. 200, creativity and. tM- 10'2 divergent thInking teal sttars and. 94 .95 kturations and giftedness33 143 eminence and, 119-20 Oft-level testing, 1113 extreme precncta. 241. 295 .247 grades and. 60-67 (11118)0g Swinunrrs. 261 group differenies, 411-43, 502 7 t higensr. 96- 97 ()twin of Spinel, The, '214 language otquisinon and. 221 handicapped modems, 281. 2$2 leadership and, 130-31 muskalay and. 194 Parents. 63-67, 318. 328, 132. 133, 334. 3:45. 337 I'Lagetian stages and, 229 31 and at tnevement. '2147-418. 2449. 279- 64). 1111) and deselopmen, of wilted children. 2514. A7 pro essional tutus and. 135 -44) And early learning. 217 1$ Khoo! performance And, 544 And early reading, 221. 223. 224 25 lerman's study. 216 of literary, prodigies, 258-59 twin studies., 10-.33 mathematically gifted, 312 underachievement And, 273. 275. :4484 variability In worm. 29-34) id musical prodigies. '254, 256- 57 verbal ability and. 222-27 and musical talent. WO IQ mu. 11. 15, 42. 51. 53. 514,60. 74. 81. 99, 114. 151. ,o1 world-class Addries. 260-61 Probed, Individual .4inonentent Teml t P1 A 14.162. 17ti. 271.1 153. 161. 044, 2*214. 246, 2446, 2117. 282. 2143. 293. Prer relationships. 237- 39 319 errinans tests and, 98-102 Physkal hanibraps. 285-86 discrepancy with Achimenient. 288 289 Piagei, jean, II. 47. 53. 54. 227, 52, 241 group differences, 301, 106 Pugnian, 153. 166, 225.r27 - 12 reading achievemem and. 135-57 Placenieni, 1423-324. 327- 32 schokutx aptitude. 70 .72 Practical intelbgeme, 57 Precocious development. 243,- 264 Preschool gifted. 216-$ Prim)" Measures 4 Must, .4whattnn, 143 e:luntopher, 3n Problem, stibmg. 52. 57, 58. 85. 94, 144. 1711, 253. 288. ensen, Anhui. 30, 31. 35, 43. 50. 225 331). 357, 3314 ohm flopkms l'niverms.1 he. 222. 247, 31) Pioryos.oriented, 337, 338 9 3 390 Production of language. lot hticluttne languagr) Speech, 158-60 161, 221,222. 230 Sputnik, 9. 73 Priv %dive language, IMi, 158-60. Standard deviation, 288, 324 Pruductivethirdong, 6, 14) Mandard Error of Measurement (SEMI. 21-22.75 Pnatinct-ortented, 337. 339-44) Staajarif-amet hisriligrare Scale. 5, 11. 75, 151,156. 721. for Verbally (kited k'ou014PV(iV). 167, '222 sr 3(km:4:hadn't), 72v, 222. 246. 356 Mathessaticat Abhor" Sternberg, Robert, 37 170 Sin ng'(ple44 farms' fauestiory, 25'2 Psythomeirk. 19. 39. 48. 52. 96 Study of Mahannawalry- Precocious l'oulhaiMPV), )1, Psychomotor ability, 239 175, 177-8, 248-53. 311 m dance. 205-6 Sun ide. 235. 240, 245 types of giftedness. 6 mid Plutaluttt Assesomett(S)MP4141. Public Ls*, 94-142, 163 Slum VI/ahoy/torsi

ELMuth*. 46, 47. 150. 151. INS. 248 au, 'revenge. 247 systems, 302 1 ask tornmtunent, 4 sm. 40 leak hen' expectations. 271- 73. 309,310 retinal players. 260- 61 lerman, Lewis, 11, 39, 215-16. 219. 234,528 Rapid arithmetic cab atom. 262 rest 59, 153, 276 RAPVIft Protect. 165 Rratimg tibrket l'ITRM. 162 Raven's Progressive Manses. 153, 24$ reu 4 E rbecitrik. Reading comprehenuon. 67, 99. 149, 157. 236. 126 interests or gibed, 2, 61. 54 Recepine laNuage, 156-51, 222 Aunts of defmniorn. 3. 941, 117, 227, 251 Refenal. 40, 65-79. 323, 44. 331 I in-tante. E. Paul, 81 Rena, Assam:41m Telt 913, 100, 1411 97.98. 190 Reasulk-Marinan Stales /at Ruling ndiavutralt two artiou ororue le46 4 Crranve Thinking, suur student,. 95 Juin ouches, 30-33. 43 Renoir Cikt's Man4411.1114,04,6 'sink. 234 Rhodes Scholats, 64 1' mkt& hirsrrnem. 56. 61, 65, 266-29:i,31)7-11. 119 Robinson. tiancs. 334 Underathievers, 56, 5t4. 64. 164. 219.'226.'239.266,213 Rural, 296. 305. 3/5. 314 19 IdrnIas, 40, '298. 3111. 503, 305 l'tban. 296,105.115, 114. 315. 116 Science aptitude. 152 (SAT). 70. 141. 163, 175. 247. 248 NikatairitAptutir 7n6 Vertu), 160, 164. 1147. 187. 169, 22),228,239.281, 2143, Sthotaxtu-4F6tudr7r.tAfaMmatttet(SA I -14), 131, 174. 875. 223. '247. 2414. 250.111 254 -145 179, ISNI,V). 156, 157. 175. 223. serhal abthly. 1444. 154). 151. 152. 153. 115. 1518. Seheficsto AptitudeIrit Verbal 246, 307.7. 8115, 306 250 s (-Thal pietism. 156-67. 219- 2'26 Sell-actuabiaoun. serhil skins, 202 Sethioncepr, 235. 2314, 2714. 290 willaItest4. 34, 46, 47, 51, 52. Mt 65. 75.97 246, 281. SEAL bee Standard f inn HI Mrasuirmeint 162.163 3)1 Serpinittal linb et Eduratranal Propo IS 1.P), tsual and perlin ming Aft. Sex, 127, 13!.. 142, 2117, 1116 [akin. 152- 211 and underat hovernent. 307. Pl, 519 ispes of giftedness, 7. 7. 16 Sexism, 307 Sex-role sterrompe. 204.167 Visual tritium tnents. 282-83 Suigk- factor thrones. 4$. 49-39, Vocational interests pie (:4reets) Single-parent home. 3114 Social annbutes, 3, 25 Wei hide", liavid, 31) Soca] development. 233 40, 114 Wrytt.kt trartivrner Stair fen,Wren, 75, 150, 2213 . 23+ inietattroin. 233-14 251. 126 Sorra) mobility., 299-3481. 320 Wei hsler S4r. On Wettish-, ImeIligeinr Slale fin(.101- Sotistiration, 313-4 Socioeconomic status ISES). 19, 33. 34, 36. 37, 41.55, 58), them 297. V9, N "WI trgsar PrIrrotri Tot, 97 61. 67. 127, 133, 135. 216, 21. 269. 279, Witir Bong, .ithunrinen4 /to 1W/1.4 1) .1511 176 Nut 115 Wit/ oti 4 NhowsiIntriltgrierr.191 sot cit dal, A. 394, 34I6. 119 WISE..14 toeWrtA4irt Intragrtur ttoOuldrrrn !axial rank. 46- 47 ti ondrorl frOntunt Prorbroduratunual (inner,. 75.7i, 162, 176, 54/MPA, 44) 276 Spatial. 283 oii&inilt sidingMaiten 104. 16'1 Spatial atalus, 152, 153. 112 Working I lAss. 23. 40. 119, 27172. 294). 297 '6)1,103 Spatial skills, 175 Spec ilk training thsaitited4.er 1..-At ttlng 014.4nItut». Wittier) tompirtnam, 1,58-434t

4 00 391